jro o DEFENSE kDVANCED RESEARCH PROJECTS AGENCY A HISTORY OF ARPANET THE The First Decade 4 1 i C % April 1981 DTIC ELECTE SWPt'ovo dltT 1'o PLtLiT o Prepared for DARPA by -- ranek and Newman Inc 1400 WILSON BOULEVARD ARLINGTON VIRGINIA 22203 S82 Li 06 10 044 Report No 4799 A HISTORY OF THE ARPANET The First Decade AMOVED 'Cl YFUBLIC RWX I'0 rsGAZ SaL UUI IMID -DISTRIBIJ 1Of ii DTIC TAB JuStifiee April 1981 DCst Ibu tJ speci Report No 4799 Bolt Beranek and Newman Inc Table of Contents I EXECUTIVE SUMMARY II 'N 'HE ARPANET PROJECT - o I-I OBJECTIVES AND RESULTS' II-1 1 'rogram Objective and Technical Need 11-2 1 1 1 2 1 3 1 4 11-2 11-5 11-7 II-7 Defense Program Addressed State of the Art at Program Inception Specific Technological Problems Addressed Expected Payoff Time-Frame Costs 2 30rogram Description and Evolution- II-10 2 1 2 2 2 3 II-10 11-12 11-19 Program Structure Major Technical Problems and Approaches Major Changes in Objectives and Approaches 3 3 Scientific and Technical Results and Accomplishments 11-21 3 1 11-21 Results of the Effort in Relation to the Program Objectives Technical Aspects of the Effort Which Were Successful and Aspects of the Effort Which Did Not Materialize as Originally Planned 3 2 4 1 4 1 4 2 5 5 1 5 2 5 3 5 4 6 Li 't Applications and Considerations 11-23 for the Future 11-28 Conclusions of Technical Feasibility Recommendations on Additional R D Requirements and Opportunities 11-28 II-28 t54rogram Impact and Assessment of Technology beveloped Service Use of Technology Impact on Non-DoD Programs Applications of the Program Results Advance in the State-of-the-Art Bibliography of Reports 11-31 11-31 11-31 11-32 11-32 11-33 Report No Bolt Beranek and Newman Inc 4799 III A HISTORY OF THE ARPANET PROJECT 1 II-12 111-2 History Background 1 1 1 1 1 The RAND Study of Distributed Communications Networks The Lincoln SDC Experiment The NPL Data Network 1 1 2 1 1 3 1 2 1 3 1 4 111-5 I1I-5 II-9 III-11 The Events of 1967 and 1968 Key Aspects of the RFQ Chronological Development 1969 to 1975 111-14 111-18 III-24 1 4 1 -'The Groups and the Key People Management and Administration of the Network DARPA IPTO DSS-W RML and DECCO The Network Analysis Corporation The Telephone Companies Bolt Beranek and Newman Inc The Network Information Center The etwork Measurement Center The Network Working Group and Host Involvement 1 4 1 1 1 4 1 2 1 4 1 3 1 4 1 4 1 4 1 5 1 4 1 6 1 4 1 7 1 4 2 1 4 3 1 4 4 111-26 111-29 111-32 111-35 111-36 111-39 111-45 Initial Subnet Design Subnet Development Host Protocol Development 1 4 4 1 1 4 4 2 1 4 4 3 1 4 5 1 1 4 5 2 1 4 5 3 111-48 111-53 III-58 Host-to-Host Protocol The Evolution of TELNET The Evolution of Other Host rotocols 1 4 5 oNetwork 1 5 1 6 111-25 Growth -- 111-61 111-64 111-68 A Summary 111-72 Traffic Growth Topology Hosts 111-72 III-76 111-91 The Impact of the ARPANET Maturity and Handover to DCA 111-94 111-104 6 6 S o' o 'o oo o Jo o i o ' o o - ' i I ' ' o 'o oiii Bolt Beranek and Newman Inc Report No 4799 111-106 2 Observations 111-107 Social Issues 2 1 111-112 Technical Lessons 2 2 2 2 1 2 2 2 2 2 3 Terminal Handling Reliability and Fault Isolation Maintenance Management III-113 111-119 III-122 Appendices A A 1 A 2 B A-1 BBN ARPANET Bibliography A-1 A-3 Technical Reports Published Papers Selection of ARPANET Logical Maps B-1 j I i I I iii wan 1- v-Ins g Report No 799 CHAPTER I Bolt Beranek and Newman Inc EXECUTIVE SUMMARY - has 4 u 1 Ji - - 1 Bolt Beranek and Newman 4799 Report No Inc EXECUTIVE SUMMARY In fiscal year Sharing Computer 1969 a Networks DARPA program was initiated entitled Resource The research carried out under this program has since become internationally famous as the ARPANET This DARPA program has created no less than a revolution computer technology and has been projects ever undertaken by DARPA extensive changes in well as in sectors the one The in of the most successful program has initiated the Defense Department's use of computers as the use of computers by the entire public and private both in telephone far-reaching the United States and around the worlu the telegraph and the printing effects on human intercommunication Just as press had the widespread utilization of computer networks which has been catalyzed by the ARPANET project represents a similarly far-reaching change in the use of computers changes set in by mankind The full impact of the technical motion by this project may not be understood for many years In 1975 the ARPANET was successfully transferred to the Defense Communications Agency which has operated time Defense Advanced Research Projects Agency DARPA I-2 it since that Repor CHAPTER II Ii Beranek and Newman Inc No 4799Bolt THE ARPANET PROJECT - OBJECTIVES AND RESULTS iI 5 i 1I-1 Report No 4799 Bolt Beranek and Newman Inc PROGRAM OBJECTIVE AND TECHNICAL NEED 1 1 1 Defense Program Addressed The DPRPA develop program techniques computers in had and the obtain following objectives experience on 1 To interconnecting such a way that a very broad class of interactions are possible and 2 To improve and increase productivity through resource sharing establishing It computer research was envisioned that by a network tying computer research centers together both goals would be achieved In fact the most efficient way to develop the techniques thought be by involving the research talent at these centers in to prototype activity permitted groups of needed Just for as hundreds an effective time-shared of computer individual users hardware and software resources with one another that networks connecting dozens of resource sharing between thousands of virtue of being time-shared network it systems to was was share thought such systems would permit users Each system by could offer any of its services to another computer system on demand The most important criterion for the type of network interconnection desired was that any user or program on any of the networked computers be able to utilize any program or subsystem available on any other computer having to modify the remote program 11-2 without I' Report No 4799 Bolt Beranek and Newman Inc This objective was an entirely new and different approach to an extremely serious problem which existed throughout botn the Defense Department and society at large computer centers in the many hundreds the private and almost completely autonomously public sectors Each computer center is forced to recreate all the software and data files that it to utilize reprogramming In many of software cases this involves extremely consuming DARPA fiscal duplicative efforts creating and year more 1969 than maintaining at language double the the software standards standardizing the types of hardware translation between complete computer for national and This and time that such costs of There had been other problem computers languages approach had some value and utility costly estimated completely different attempts to address this attempts the wishes or reformatting of data files duplication and redundant effort is In of Defense Department and the many other thousands of computer centers in operate The attempts such as attempts at at automatic Although the problems each of such trying to share computer software resources or files was truly enormous In addition the scientific certain to the general problem shared with the rest of community special problems the Defense DepartmeTrt also having to do with training Military personnel trained to use one manufacturer's equipment must be trained again to use another's 11-3 Machines faces procured often from j Report No 479' Bolt Beranek and Newman inc different manufacturers require as many different programs as there are machines of training that could military personnel common must utility be training thus inhibiting positive transfer accumulate through Those data files and the rotation programs which of have to many military organi ations and installations stored different user created machine and maintained Military systems separately at each interconnected in a distributed intiractive network obviate such constraints Another objective of th of specialized centers in the It computers program was to permit to area it compiling list Making such machines establishments would recent improvements would become most cost effective to design and construct computers e g linking the many general purpose computer was thought that with the then hardware the efficient processing available to at specialized tasks and information retrieval all significantly the increase computer research the capability at these other centers This program was addressed et no less then changing the of computers by the entire Defense Department intended that the use of such a resource sharing within and computer across A was clea ly would permit the military services and throughout the Defense research community 11-4 network It use Report No 1 2 4799 Bolt Beranek and Newman Inc State of the Art at Program Inception By the date of the program plan in the specific technologies required individually been achieved in the late 1960s some form However Air Line reservations systems there of for a computer network had For example been many connections of phone lines to computers system most there e g had the SAGE and time-sharing systems had been only a very small number of attempts to connect computers together for the purpose of experimenting with the sharing of resources o In the early 1960s an attempt was made to link computers together at the Western Data Processing Center at UCLA for the purpose of enabling similar computers to perform load sharing A similar experiment was also performed at Bell Laboratories and achieved reasonable success for several years o A number of networks were constructed purpose of inventory message control reservations handling system networks The for the primary including a Westinghouse and several air line SITA network for air line reservations was surprisingly advanced in concept in the mid-1960s generally not known in but details about the U S techniques used SITA were computing community for In any case the such message systems were special 11-5 Report No 4799 Bolt Beranek and Newman Inc purpose in nature and were not readily transferable the general o into area of inter-computer communications A direct progenitor of the ARPANET was an effort made in the mid-1960s to achieve a coupling computing expertise computer This effort led to a between the Laboratory Q32 and modifying and at the SDC time sharing academic operation of the SDC Q32 phone and demonstrated between line the the connection TX2 at relative systems to Lincoln ease permit of network interactions Aside computers from with the communications networks had been theoretical technical considered point of view problems circuits in Communications was work done by Donald Physical Laboratory In sum in fiscal the notion of computer number of in Davies in England in Corporation been places in a the early 1960's and others from at study On Also of note the National at the time of the initiation of the ARPANET program year 1969 attempted in ever been made to put many of the requisite some form them ideas bits and had been pieces but no significant attempt had together into a resource sharing computer network 11-6 a the mid-1960's considered and many of the requisite technical had interconnecting Of particular note was work done by Paul Baran and others at the Rand Distributed a of 4 Report No 1 3 4799 Bolt Beranek and Newman Inc Specific Technological Problems Addressed The technological problems of building the ARPANET can be considered at many different levels of detail At the top level there were really two problems 1 To construct circuits a and subnetwork switching characteristics consisting of telephone nodes whose reliability delay capacity and cost would facilitate resource sharing among the computers on the network 2 To understand procedures design within connected computer and implement the the operating protocols systems of and each in order to allow the use of the new subnetwork by those computers in sharing resources Within these two major technological problems of course a large number of sub-problems engineering of the phone circuit connections network were including the the topology of the the selection of switching node equipment line disciplines to work through phone line errors problem there the design of the routing and many others I- 1 4 Expected Payoff Time-Frame Costs The goals for the ARPANET project were envisaged a significant eventual impact on the use 11-7 very broad and of computers Report No in 4799 Bolt Beranek and Newman Inc both the public and private sectors these long range goals initial DARPA However visualized some in addition to quite specific payoff in the form of improved productivity of the DARPA research program itself and a resulting cost performance benefit to the services from DARPA research number of In fiscal 1969 a computer research centers throughout the country were supported in whole or in part by DARPA's Techniques year Office IPTO The Information installation of Processing an effective network tying these locations together would substantially reduce duplication and improve the transfer of well as develop scientific results the network techniques needed by the military The research output of these projects was inportant to all Services this and as it was expected that output substantially increased for the same dollar cost if three could be a portion of the funds were utilized for the network In addition initial payoff was anticipated in the form of technology transfer from the ARPANET project in three ways o By dissemination of new scientific knowledge through conferences and the appropriate literature o By transfer of management of the carrier and the resulting Laboratories Office a of of common ARPANET Education NSF-supported universities various user groups supported by the NIH 11-8 to availability services to other groups such as Regional ARPANET and Report No U By 4799 Bolt Beranek and Newman Inc adoption military System of groups Support affiliated wth the network technology by specific such as the National Military Command Center and it e g 11-9 other CINCPAC military CINCEUR centers and MACV Report No 2 2 1 Bolt Beranek and Newman Inc 4799 PROGRAM DESCRIPTION AND EVOLUTION Program Structure With the initiation of the ARPANET fiscal year 1969 paths of effort select the DARPA began work in 1 to obtain system contractor overall design and 3 research program community to in early earnest on three parallel the network for the switching nodes and the initiate for plan circuits efforts resource within sharing 2 the to DARPA experiments and specialized network support DECCO was able to handle all the the contractual common carriers for circuit leases kilobit circuits used in from the ARPANET were Most of the required 50 leased carriers such as General Telephone In addition for in special point of contact and DARPA contractor locations and the of internode circuit The AT T selection connections terminations was and right To help solve this particular contracted with the Network Analysis Corporation developed DECCO certain networking arranged DARPA network which system of network node therefore the a specialized topology problem and represented a difficult theoretical own through AT T long lines greatly facilitated the interactions between the location with but a small number of circuits were leased from other AT T a details problem in problem NAC its DARPA NAC had analysis tools and via this DARPA II-10 hal' 'd A'o- o - - - - o- oo I I Report No support 4799 Bolt Beranek and Newman Inc such tools were refined NAC's advice on topology was sought through the various stages of ARPANET growth A competitive procurement was planned for the selection of the contractor to design the switching nodes and act systems contractor Bolt Beranek and An RFP was prepared and issued in Newman Inc BBN contractor for the design of the nodes BBN January Over the 1969 until life subnetwork the of transfer the to and and maintenance of the July 1968 ARPANET DCA in switching program from July of 1975 subnetwork serving that role under the aegis of the the for the switching node served as the systems contractor for the design operation general was selected as the systems subcontracted with Honeywell hardware itself as Defense BBN implementation BBN is still Communications Agency DCA The research groups receiving DARPA IPTO support then began considering procedures the and design in turn various host computers in specific responsibilities and implementation of protocols computer program modifications order to use the subnetwork were arranged in and the Several UCLA was specifically tasked to develop and run a Network Measurement Center with the objective of determining the performance of the network SRI was specifically tasked to develop and operate a Network Information Center with the objective of collecting lI-11 information about the Report No network 4799 Bolt Beranek and Newman Inc about host resources and at the same time generating computer based tools for storing information hoe Network various were research protocols accessing for to contractors Group sites was reach about intercommunicating Working host pursued of the over interested eventually a collected some rather ad agreement between the appropriate the host subnetwork individuals The from the rather informally encouraged by DARPA After a time this Network Working Group became and that Beyond these two specific contracts mechanisms various and semi-official approval the forum authority for for the discussion of and issuance of host protocols to be implemented by the various research contractors Progress in this area was rather slow for a while but with time this mechanism eventually was successful in establishing effective host protocols 2 2 Major Technical Problems and Approaches In a program of this duration and complexity it is not difficult to identify many dozens of important technical problems and approaches to those problems We here list a few of the problems which were most technically challenging in the early few years of the ARPANET program A few additional major technical problems will be listed in the next section on Major Changes and Objectives 11-12 Report No o 4799 Bolt Beranek and Newman Inc TOPOLOGY - For any network of dozen nodes formidable Assuming an obvious problem that is this type early recognized topological links among delay functions the component and reliability typically non-linear constraints a optimization N the number nodes large the links are usually available in and even and quite the node locations are known of ways of arranging M bandwidth with is very- discrete sizes cost structures functions time are all The design may be subject to many maximum including or average time delay average or peak throughput requirements and reliability requirements The usual goal of the optimization is to provide a network design that meets all constraints at the lowest cost design an elaborate optimization approach a The approach to this It is problem was to computer program to assist in the not possible to use an exhaustive and instead the approach used was to generate 8starting network transformations to and the then topology to in perform local order to reach a locally optimum network If this procedure is repeated with many starting networks and if the resulting locally optimum networks are evaluated feasible it is possible to find a solution with costs that are close to optimal 11-13 V Report No o 4799 Bolt Beranek and Newman Inc ERROR CONTROL - A critical sharing computer necessity network was for to a resource provide reliable communication and one component of such reliability an ability to work through expected phone line the errors on the 50 kilobit circuits was to design transmitting circuit in special and The approach checksumming receiving end was taken hardware of each 50 at the kilobit As part of the switching node the network transmission procedure checksum appended to every packet of information to is a powerful 24 bit cyclic the at the receiving node Then be transmitted checksum is recomputed in hardware and compared with Lhe transmitted checksum If there is error an back to an error no packet will be acknowledgment with its own checksum the transmitting acknowledgment node If there sent and the is no is sent is retransmitted 0 3 HOST INTERFACE design of an An important issue was the proper interface between the switching node and host computers of many different types It to allow a logical switching match between the is important node computer word length and the varying word lengths of the host computers routines of the and also switching 11-14 to allow node and the input output the input output i Report No 4799 Bolt Beranek and Newman Inc routines of the host computer to service the interface cooperatively burden or without tight placing an undue processing timing constraints on either machine The approach taken was to design a bit-serial which could require logically that specialized each small stop after any bit and to then host computer hardware build itself obtain unit called a special interface which would be installed in a interface the host to serve the network connection Thus a host machine a logical and electrical convention was specified by the switching network in to all a manner intended to minimize overall trouble hosts then each host was required to meet that standard both in hardware and software 0 SWITCHING NODE PERFORMANCE network was a such a way i-emote resource particular this sufficiently reliable sufficiently A central goal of the to provide resource sharing between remote installations in employ - great required that a local without that degradation the have sufficiently capacity user could In subnetwork be low delay have and have sufficiently low cost that remote use would be as attractive as use These objectives translated into a major technical problem for the switching 11-15 node itself local to provide low Report No 4799 Bolt Beranek and Newman Inc delay and high capacity at modest taken was to cost The approach select a mini computer for the switching nodes whose I O system was very efficient and very to write carefully tailored programs in machine language to optimize the capacity and the low delay of the data path in the switching node minimizing the Great operating attention time of was paid to the inner loops of these programs o REMOTE CONTROL - A special and new problem was posed by the need to put dozens of small identical mini computers in the field in an environment where the host connections to those computers was somewhat experimental and where the programs in the mini computers were under month development and were changing from month to It was important to be able to do software debugging modifications without themselves the and costs of the new host connections installation and debugging difficulties of new The approach taken was to an technology entirely management new A Network Control of Center program programs associated having manned sites remote was of with develop computer established for the subnetwork and software was developed which made it possible to examine or change the operating software in any node of the net from the central network 11-16 control Flow_ Report No Bolt Beranek and Newman Inc 4799 it made approach This center in network from a central place in an hour or twc In health to addition the each node reports the state of its central place periodically and provides information maintenance activities ROUTING - In a non-fully connected network an important on which to base debugging and o issue node each complete new versions of the software to the to possible problem the is which each node by process decision decides how to route information to reach any particular This is destination a including fixed and there are many different approaches routing random problem theoretical difficult centrally controlled routing routing The and various forms of distributed adaptive routing approach taken was to use a distributed adaptive traffic routing information from loads on the the and proper the face of in local of basis globally message each for varying input traffic failures nodes adjacent path instantaneous estimates which algorithm line or nodi Each IMP keeps tables containing an estimate of the output circuit corresponding to the minimum delay path to each other IMP the delay Periodically routing estimates to receives and a corresponding estimate such a its each neighbors message II-17 IMP sends its it whenever updates its of current an IMP internal Report No 4799 Bolt Beranek and Newman Int estimates is Thus regularly information about changing propagated throughout the conditions network and whenever a packet of traffic must be placed on the queue for an output circuit the IMP uses its latest estimate of the best path o HOST PROTOCOL - In many ways a computer host computers users as the telephone system is is to to human a transparent communication medium in which even after the caller has learned how dial network it is language as still the communication necessary person to as host protocol protocol which is insert that called to occur to in dimes and he speak the same order for The common language is and the problem is useful referred to design a host sufficiently powerful for the kinds of communication that will occur and yet can be implemented in all of the various different host computer systems The initial approach taken involved the development of a piece of software called a Network which a host would reside in processes within a network this through host would Network primary function of the NCP is break connections computer communicate Control Program such with Program that the The to establish connections switch connections U1-18 Control and control flow Report No 4799 Bolt Beranek and Newman A layered approach was taken such that more procedures such as File Transfer Procedures on top of simpler procedures in Inc complex were built the host Network Control Program 2 3 Major Changes in Objectives and Approaches The ARPANET development was an extremely intense activity in which contributions scientists in technical were made by the United States problems many Thus already times However in during the THE TIP - The IMPs were received those years addition several more were introduced once the initial o to early the best computer almost all of the mentioned attention and the detailed approach several of major major continuing problems changed of the ARPANET effort changes in objectives network became operational initial nodal switching units called intended to interconnect computers and high bandwidth phone lines access network was via terminal connections to to the the hosts themselves there was a access to At the outset After a time it population cf users for all terminal became clear that which the network was very desirable terminal but who were not conveniently able to access the network via computer Thus a new nodal switching unit Interface Message Processor 11-19 or TIP was a host a Terminal defined to Bolt Beranek and Newman Inc Report No 4799 the purpose of an IMP plus an additional function serve of direct terminal access This shift resulted the in design of a TIP which really was a tiny host embedded in a switching node itself connection of up to 63 terminals to the and a local the direct asynchronous character-oriented node The TIP became the switching nodal switching unit of choice was permitted often even where there host computer this allowed connection of both hosts and terminals at that location directly to the network 11-20 1i II Report No 3 4799 Bolt Beranek and Newman Inc SCIENTIFIC AND TECHNICAL RESULTS AND ACCOMPLISHMENTS 3 1 Results of the Effort in Relation to the Program Objectives In some cases a major program objectives in a single instant can be seen to reach its a mushroom cloud at Alamogordo or Armstrong stepping onto the moon In other cases like the ARPANET an equally important objective may be reached with equal success but the event must be observed in and over was to a longer period of time develop interconnecting computers of interactions are entire techniques in a more complicated way The first and ARPANET objective obtain experience on such a way that a very broad class possible Not just techniques but technology of packet switching has been developed an and an enormous body of experience has been developed on interconnecting computers to allow a broad class of interactions A second objective which has also been attained was to improve computer research productivity through the development of computer resource sharing Another objective was to permit the linking of specialized computers to the many general-purpose computer centers major the Several specialized computers have been linked over the ARPANET to main general purpose computer successful fashion 11-21 centers in an extremely Report No At 4799 a Bolt Beranek and Newman Inc technological level the overall objectives were to construct a subnetwork whose reliability capacity computers and in cost the network and to understand permit consisting Hawaii to Norway of such over design and implement resource 50 nodes sharing resource sharing computers were successfully understood demonstration was among the connected designed and implemented of initial net operation with four nodes a time scale extremely close to the incremental time anticipated in the program that was being planned the However as the ARPANET became obvious after about two years success that had never been anticipated at the outset more than 50 nodes over a geographic area from Hawaii was certainly the costs in close not originally anticipated the first the scale of the to network a growth to 'o Norway In similar fashion two fiscal years of the program were to anticipated costs of decisions were made to take advantage of this success and to grow the net size a and the necessary and later the extension to 19 nodes took place on that Such and stretching from was successfully constructed host protocols to allow The characteristics would facilitate resource sharing among the the host protocols to subnetwork delay very but as decisions were made to expand the cost original program plan 11-22 no longer followed the Report No In 4799 Bolt Beranek and Newman Inc short successful and initially the far results more stated of than the effort adequately were met eminently the objectives The success of the program far exceeded even the most optimistic views at the time of inception 3 2 Technical Aspects of the Effort Which Aspects of the Were Successful and Effort Which Did Not Materialize as Originally A representative list of the aspects of the ARPANET programs Envisaged which were technical successes o Powerful follows computer-based optimization were techniques developed for topological and the choice of ARPANET topology was made with the aid of these tools o The carriers successfully provided high reliability 50K sec circuits o A subnetwork including nodal switching IMPs and TIPs was constructed whose performance reliability and cost did facilitate resource sharing o The ARPANET provides adaptive routing reliably e g of 1 local in a convincing algorithms can demonstration be made to perform a globally correct manner in failures efficiently 11-23 e g that the face adapting to Report No Bolt Beranek and Newman Inc 4799 changes in network the circuit of variety distances without excessive internode and bandwidths and and accurately a accommodating e flexibly quickly complexity and overhead o The ARPANET has demonstrated that it is possible to build a large operational network in such a way that the of component failures are localized rather than effects crashing or portions or otherwise all making of the network fail in the ARPANET and network use for only large non-operational A node or a host can prevented be will the few users directly connected to that node or using that host o The ARPANET has confirmed the delays store-and-forward which networks are which incurred in theoretical the low when packets compared computers hosts to result that can achieve the delays which are using the network o The ARPAMET has demonstrated constructed so that nodes be added or that a network lines traffic can be and so on can deleted without major upheavals with each addition I1-24 Report No o 4799 Bolt Beranek and Newman Inc The ARPANET has demonstrated that it network to control and is possible operate itself for a for hours at a time without explicit control from a control center o The ARPANET monitoring has demonstrated maintenance new and techniques debugging for The nodes in the ARPANET typically operate at sites where there is krowledgeable person available automatically report their itself and locally status via to a network monitoring center debugging of The the and no nodes network maintenance both the software and hardware typically carried out again via the network from are the monitoring center o Possibly the development most of the proved successful -host computer difficult task ARPANET to make systems of was a in the attempt -- number varying varying operating systems within a type undertaken of which independent manufacture single the and manufactured communicate with each other despite their diverse characteristics o A set of host protocols was host organizations hammered and resulted in out between a layered structure of host protocols that did facilitate resource 11-25 the sharing -' Report No o 4799 -wla oL---o Bolt Beranek and Newman Inc The ARPANET was successfully transferred to the Defense Communications Agency Inevitably tractable there than had were been technical hoped and areas some which proved less kinds of network utilization developed more slowly than had been anticipated Problems of routing the subnetwork flow control turned out theoreticians had anticipated and congestion control in to be as difficult as and the algorithms had to be modified more than the implementors had anticipated Luckily the improvements in the algorithms managed to stay slightly and ahead of therefore represented the growth in network size and traffic difficulties an impediment with to the algorithms resource sharing never on the network It proved more difficult to agree on adequate However host protocols the eventual than design had of the specification of been originally hoped the host protocols was eminently successful and provided a strong base for resource sharing The development I k of Information Center NIC a widely-useful proved to be a on-line difficult Network project Although very interesting and important computer-based 11-26 tools Report No for 4799 Bolt Beranek and Newman Inc information handling were developed the actual timely collection and on-line publication of detailed of resources intractable developed at the various Further while the rather host sites computer elegant effective for wide scale remote use descriptions proved tools nearly that were were not so clearly cost Eventually the goals of the NIC wer-e substantially reduced and at the time of the ARPANET obtain transfer detailed to DCA it inform3tion was about generally necessary to the resources at a particular site directly from the site Many different kinds of resource anticipated at the inception of these the ARPANET use were program Of remote use of one program by another was one of the more distant goals sharing has the sharing indeed While some of this kind of resource taken place such Use has grown more slowly than other network uses t1 11-27 Report No 4799 4 Bolt Beranek and Newman Inc APPLICATIONS AND CONSIDERATIONS FOR THE FUTURE 4 1 Conclusions of Technical Feasibility The ARPANET achieving reliable communications turn project by high proved the performance means technical cost effective network digital of packet switching technology and the technical feasibility of operating a computer feasibility of based on this technology resource in sharing The ARPANET project also proved the feasibility of achieving closely knit communities of technical interest over a widespread geographic possible that this social feasibility area it is demonstration is as important as the many technical feasibility demonstrations 4 2 Recommendations on Additional R D Requirements and Opportunities It is clear that packet direct application to command research exists opportunity switching and in command and control applications control networks have a very and a significant attempting to investigate such DARPA is already proceeding to develop testbeds whereby packet switching technologies and related technologies can be experimentally employed cooperation with one or more of the military services a clear requirement for improved There in is command and control and the packet switching technology developed in the ARPANET major opportunity to make progress towards this goal 11-28 other provides a Report No 4799 At a Bolt Beranek and Newman Inc deep technical level the proper design of host operating systems for efficient and cost effective connection networks is still in the development opportunity connections were more detailed area In the add-ons of ARPANET and it investigation of significant to research and project the network is clearly time to mount the how best to accomplish such connections The techniques for remote control of computers in the field developed within the ARPANET project are probably more broadly applicable to the management of computer resources in other areas of the Defense Department These remote management techniques represent another opportunity which has grown out of the ARPANET experience It was earlier indicated that it was not practical for the Network Information Center in the ARPANET to keep current on-line detailed descriptions of program resources available host community and development This within the difficulty in turn represents a research opportunity for the future Specifically research and development is required on how to properly describe computer programs descriptions such available resources of the ARPANET and how that it to create for such will be easier to create compendia of In other words the standards basic despite the resource 11-29 sharing great problem success still Report No 4799 Bolt Beranek and Newman Inc represents a fertile area for research and development the networking technology itself is given Now that attempts at description and documentation of host resources might more easily yield to a research and development effort IJ S 11-30 Report No 5 4799 Bolt Beranek and Newman Inc PROGRAM IMPACT AND ASSESSMENT OF TECHNOLOGY DEVELOPED 5 1 Service Use of Technology The ARPANET technology is being succe sfully transferred to the rest of DoD Not only was it possitle ARPANET itself to the Defense Communicationr to transfer Agency Code the 535 but the Defense Communications Agency has already embarked on the procurement of of the future -- packet its primary backbone major communications system AUTODIN II switching in based technology beyond this very major involved -- very developed step all investigations three of specifically in the services on ARPANET are the Even actively packet switching technology for their specific needs 5 2 Impact on Non-DoD Programs In just the very short ARPANET Tymshare since the inception of the this technology has already resulted in the formation of a new industry packet time a private sector development switching networks Two of corporations are currently marketing packet switched value added Telenet and communications to the general public as common carriers under the Communi' ations act f of 1934 This program therefore has also had direct and immediate exploitation in the commercial Now DCA Code 252 December 1981 11-31 sector In addition Report No all 4799 Bolt Beranek and Newman Inc over the worlc new communications systems are being designed and built to take advantage of the packet switching technology demonstrated by the ARPANET project At least three countries Great -- Britain France and Canada have major national PTT- sponsored packet switching networks either already under -- operating or development and many other countries are actively pursuing this technology 5 3 Applications of the Program Results The most specific result of course been the ARPANET the itself ARPANET and the Program ARPANET itself currently fully operational under the management of Communications Agency and is actively has serving the of is Defense thousands of The ARPANET program has represented a first-rank advance in individuals on a daily basis 5 4 the Advance in the State-of-the-Art state-of-the-art of communications and the state-of-the-art of computer technology provision of communications have also resource cost effective but very taken optimization The greatest advance significant multiprocessor between programs remote computer management 11-32 been in the high per-formance digital state-of-the-art place in many other areas routing sharing reliable has such as topological technology network advances protocols mail systems for and Report No 6 4799 Bolt Beranek and Newman Inc BIBLIOGRAPHY OF REPORTS There has been an enormous amount of ARPANET in networks in the particular general and about literature resource about sharing Literally an industry has been the computer formed in space of two-thirds of a decade and the amount of literature reflects this really unusual metabolism The initial May of seminal papers on the ARPANET were presented 1970 in at the AFIPS Spring Joint Computer Conference in Atlantic City and are published conference AFIPS 210 Summit Avenue ARPANET session Conference in Conference Montvale was held in the Proceedings Proceedings New Jersey at the 07645 1972 Atlantic City and these Vol 36 that AFIPS Press Another early Spring Joint C mputer papers AFIPS Conference Proceedings Vol 40 papers represent a sensible introduction of are published in These two early sets of to the early notions about and plans for the ARPANET A sizeable bibliography and index to publications about the ARPANET was published in 1976 with DARPA support Selected Bibliography and Index to Publications About the ARPANET and Hayes Inc February 1976 185 document represents the most complete ARPANET is publications provide help in but it paias AD-A026900 availa le listing Becker This about a bibliography only and does not trying to select which 11-33 of the many references Bolt Beranek and Newman Inc Report No 4799 Another bibliography would be suitable to look at in what order on very useful in helping the reader decide not also is document not important and what is what is and provide a much easier entry to 1 the literature than the large bibliographies Communications Computer Dedham Washington St Communications Wesley W Chu East Green E 47th IEEE Press 345 10017 1975 and 3 Computer Networking Blanc and York New York 10017 1974 Jr Street Lucky Cotton Advances in Inc 610 Artech House 02026 Massachusetts edited by Paul Ira W the together collect to attempt themselves papers important which but specifically the ARPANET than rather general in networks computer deal which Several volumes of reprints have been published with This 1976 revised 384 Publication Special NBS Networks Sharing Computer Resource on Literature the of U S Annotated National Bureau of Standards Department of Commerce Bibliography the by issued been has ARPANET general not just the networks in computer sharing resource of literature the Computer 2 and Robert New York W New York edited by Robert P IEEE Press 345 East 47th Street New 1916 A very hefty but fairly readable compendium with a very large list of references is Infotech State of the Art Report 24 Network Systems and Software 11-34 Infotech International Ltd j Report No Bolt Beranek and Newman Inc 4799 Berkshire Maidenhead Nicholson House 1975 England This the ARPANET but it tries deals with far more than just other current work as of 1975 to put the ARPANET in context with document reference There are three important been prepared for the Defense documents Communications have which Agency by the Menlo Research Institute Network Tnformation Center at Stanford addressed to 94025 which are specifically California Park 1 ARPANET Resource Handbook various aspects of the ARPANET NIC Protocol Handbook ARPANET NIC 39335 December 1976 2 3 ARPANET Directory NIC 36437 7104 Revision 1 April 1976 July 1976 A on textbook computer Communication Networks for special issue of networks Computers is John Davies Wiley and Barber 1973 A Communications the IEEE proceedings on Packet was issued in November 1978 11-35 Inc Bolt Beranek and Newman Report No 4799 CHAPTER III PROJECT A HISTORY OF THE ARPANET 1 I I II tt- SA Report No 1 4799 Bolt Beranek and Newman Inc HISTORY The DARPA Computer Network called consists of IMPs The IMPs short for special-purpose lines lines Interface computers The hosts are a or ARPANET as it and hosts as shown in Message Processors connected be Figure 1 are small to each other by telephone heterogeneous collection of computers HOlST IHO St IHOST _ has come to PHONE LINES Figure 1 used for a variety communications each other -- of subnetwork IMPs Lines applications through and Hosts The provide a which hosts communicate with much as the commercial telephone 111-2 IMPs system provides a Report No 4799 Bolt Beranek and Newman Inc communications other words subnetwork through which humans communicate the IMPs and lines make up a communications of which the hosts are users one IMP may have several When is data is Each host is which is hosts connected to it to be sent from one host to another consists of also called the to be sent host Successive messages its where they are IMP packets The at the sending Each connected The address specifies that is the destination are passed from the source host to broken into smaller entities at the IMP to which the destination called host is and passed to the destination host ARPANET was conceived in the middle to late 1960s as a project to be sponsored by the Information Processing of of the U S network to where the original messages are reconstructed from the The Office message are routed from IMP to IMP across the network until arrive packets host the data each of which carries the same address as the message packets they source some data and an address which host the data is utility connected to one IMP broken into discrete entities called messages host In Techniques the Defense Advanced Research Projects Agency DARPA Department of Defense of began ir' 1969 By DoD Construction the 1975 the network had gone through several stages of development and for all intents and purposes had become an operational Department of Defense computer network 111-3 r Report No In 1975 U S 4799 Bolt Beranek and Newman Inc control of the network was transferred from DARPA to the Defense Communications Agency an agency better suited to the administration of a working facility The ARPANET is a computer major communications history of the ARPANET the major design associated with it and why it is development Our why it deci3ions its purpose in here the evolution is to present the was built who helped and some of the evolutionary development important 111-4 of build minor its it ones maturity 4799 Report No 1 1 Bolt Beranek and Newman Inc Background 1 1 1 The RAND Study of Distributed Communications Networks One of the most important early studies of computer networks was performed by Paul Baran Corporation in the early 1960s and his Many colleagues at the RAND concepts central to the later development of the ARPANET and other computer networks were first described 1964 a list end of in the series of reports published by RAND in of these reports is given in the bibliography at the this subsection These ideas include the improved reliability of a distributed network structure over a centralized or star network and over so-called decentralized networks made up of a collection of smaller star networks undertaken including determine how expected to simulation survivable be a of Extensive studies were some distributed grid network after heavy node and link failures was particularly concerned with the question of percentage of networks the could to be This study keeping high a network available and performing well in the face of enemy attacks on the network# from the point of view of its suitability for Department of Defense applications In specifying closely the engineering details of what was Network called the Distributed Adaptive Message Block anticipated practical many of the developments came a full decade later In the 111-5 in Distributed Baran networks that Adaptive Message Bolt Beranek and Newman Inc Report No 4799 Block Network terminals of user-to-user a multiplexing widely station differing cryptography is low-cost micrrowave up to 1024 characteristics integrated switching technique to ensure efficiency and connects into Automatic the network Both satellite links relay systems are suggested as techniques for providing the npt- 'ork with very high data rate circuits The a packet of up to concept of a mesrav block s introduced 1024 bito of transferred in of this header Lhe network study is was which unit of data One of the most interesting aspects not only is the that a large-scale digital feasible but also highly and proposed that many of the switching functions be implemehted in hardware extremely data that 1t concluded transmission network cost-effective and reliable Baran was considering ways of making networks and so preferred simple solutions and reliable hardware where possible The following bibliography includes the entire set of eleven reports in the original RAND study as well as two published papers resulting from that set and trom two later reports annotated version of this bibliography Routing Algorithms McQuillan for BBN Report No is included in Distributed Computer Networks 2831 May 1974 11 pp 14-17 An Adaptive John M Report No 4799 Bolt Beranek and Newman Ino Bibliography P Baran On Distributed Communications Distributed Communications Networks August 1964 37p I RAND Corp Introduction to Memo RM-3420-PR S Boehm and P Baran On Distributed Communications II Digital Simulation of Hot-Potato Routing in a Broadband Distributed Communications Network RAND Corp Memo RM-3103-PR August 1964 4 9p J W Smith On Distributed Communications III Determination of Path-Lengths in a Distributed Network RAND Corp Memo RM-3578-PR August 1964 91p P Baran Precedence 1964 63p On and IV Priority Distributed Communications Overload RAND Corp Memo RM-3638-PR August P Baran On Distributed Communications V History Alternative Approaches and Comparisons RAND Corp Memo RM-3097-PR August 1964 51p P Baran Microwave On Distributed Communications VI Mini-Cost RAND Corp Memo RM-3762-PR August 1964 101p P On Baran Engineering Distributed Specificatio ' Communications and High-Data-Rate Distributed Network Switching Memo RM-3763-PR August 1964 VII Preliminary Tentative Design Node for RAND a Corp 8 5p P Baran On Distributed Communications VIII The Multiplexing Station RAND Corp Memo RM-3764-PR August 1964 103 p P and Baran On Distributed Communications Tamper-Free Considerations RAND IX Security Corp Secrecy Memo RM-3765-PR August 1964 39p P Baran On Distributed Communications X Cost Analysis Corp Memo RM-3766-PR August 1964 21p P Baran On Distributed Communications XI RAND Corp Memo RM-3767-PR August 1964 23p Summary Overview P Baran On Distributed Communication Networks Transactions on Communications Systems Vol CS-12 March pp 1-9 111-7 RAND IEEE 1964 I Report No Bolt Beranek and Newman Inc 4799 emo Motn Adaptive and R L Mobley B W BoehmCommunications SYstems RAND Corp M iqe fo47-r Distributed February 1966 78p Simulation of Adaptive L Mobley A Computer R RAND and Boehm W B Communications Systems Distributed for Routing Techniques February 1966 35p Corp Memo R14-4782-PR for Adaptive Routing Techniques on Mobley L R and Transactions B W Boehm IEEE Systems 3 June 1969 pp Distributed Communications Vol No COM-17 Technology Communication 340-349 1 4 IH 1I1-8 Report No 1 1 2 Bolt Beranek and Newman Inc 4799 The Lincoln SDC Experiment The notion of by J C R envisioned reality linking computers Thomas Corporation of Marill America commissioned by M I T computer networking and in his Network of on the of America later paper of the Lawrence Roberts the also in constructed computers that The fact Computers Thomas authored appeared in suggested late 1965 by Marill 11 Computer June 1 Thomas 1966 a Marill and pp 425-431 software The considered the and problems experimental linking report three network and II I-9 Lincoln be existing the AN FSQ-32 at Systems Development Corporation IBM 7094 at MIT's Project MAC the the Proceedings of the AFIPS three-computer report Lincoln 1966 A Cooperative idea of computer networking a were a subcontract under Technical Report No name at The study was done in available communications techniques and recommended contact study was issued in same basic Computer who played an active role in 1966 Spring Joint Computer Conference examined at Massachusetts The primary technical Time-Sharing Corporation was Lincoln Laboratory to study the concept of the Laboratory's DARPA contract report colleagues Cambridge network study and the contract was a network director of DARPA IPTO Laboratory was Lawrence Roberts and a Licklider long before the ARPANET become a Licklider was the first In 1965 into Laboratory's the TX-2 Report No 4799 Later 1966 in Bolt Beranek and Newman Inc CCA received another contract to carry out the The Q-32 and TX-2 were in linking of the Q-32 and the TX-2 linked together and the link was demonstrated Later a fact small Digital Equipment Corporation machine at DARPA was added to this network by now known noteworthy directly that The as The Experimental Network It I is Experimental Network linked host computers and did not use IMPs i Ii III-10 S o ooo ' o-'i o o o- -o ooo ' Report No 1 1 3 4799 Bolt Beranek and Newman Inc The NPL Data Network Another early development of major the network ARPANET was development undertaken Physical Laboratory in Middlesex England of system D W Network Davies as it The broad which at under the the design affected National leadership of the NPL Data was called was first published in 1967 and bears a resemblance to the network proposed by Paul Baran at RAND to the ARPANET packet-switching structure The NPL network Data and and Network was specified to be a was to have a hierarchical It was proposed that local networks be constructed interface computers which had responsibility for with multiplexing among a number of user systems and for communicating with a high level with switching network nodes The latter would be constructed connected together with megabit rate circuits Considerable detail about the NPL Data Network may be found in the textbook written -entitled D L A Communications Barber bibliography of John by two of the network's designers Networks for Computers D W Davies and Wiley published Sons papers publishers resulting Network effort follows iIi - mrlmm 1973 A from the NPL Data Bolt Beranek and Newman Inc Report No 4799 Bibliography T and P Scantlebury R A Bartlett D W Davies K A Wilkinson A Digital Communication Network for Computers Giving Rapid Response at Remote Terminals ACM Symposium on Operating Systems Problems October 1967 17p Bartlett R A Scantlebury and P K A Real-Time for Network Communication International Conference on Communications Wilkinson A Data IEEE Computers U S A June 1968 T Networks to Serve Rapid Response Communication D W Davies Vol 2 1968 Computers Proceedings of the IFIP Congress Hardware Applications pp 650-658 K A Bartlett Transmission Control in a Local Data Network Vol 2 - Hardware Proceedings of the IFIP Congress 1968 Applications pp 704-708 The Principles of a Data Communications Network D W Davies IFIP of the for Computers and Remote Peripherals Proceedings Congress 1968 Vol 2 - Hardware Applications pp 709-715 Wilkinson and K A Bartlett The T P A Scantlebury R Communication Design of a Message Switching Centre for a Digital 2 Vol 1968 of the IFIP Congress Proceedings Network Hardware Applications pp 723-733 in P T Wilkinson and R A Scantlebury The Control Functions 1968 Congress Proceedings of the IFIP a Local Data Network Vol 2 - Hardware Applications pp 734-738 Colloque D W Davies A Versatile Data Communication Network Paris March 1969 pp sur la Teleinformatique International 461-467 Scantlebury A R and D V Blake Barber A D L for Data Interface of the British Standard Implementation Colloque Data Digital of and Acceptors Exchange between Sources pp March 1969 Paris International sur la Teleinformatique 807-816 Bartlett R A Scantlebury and P T Wilkinson K A Transmission over Half-Duplex on Reliable Full-Duplex CACM Vol 12 No 5 May 1969 p 260 A Note Lines D L A Barber Experience with the Use of the British Standard Interface in Computer Peripherals and Communication Systems ACM Symposium on Data Communication Pine Mountain October 1969 III-12 Report No Bolt Berarnek and Newman Inc 4799 for the Local Area of a Data A Model R A Scantlebury and Hardware Organization Communication Network - Objectives Mountain October 1969 ACM Symposium on Data Communication Pine a Data A Model for the Local Area of Wilkinson T P on Symposium ACM Communication Network - Software Organization 1969 Data Communication Pine Mountain October The NPL Data Network Barber and D W Davies A D L Automation Laboratory On Conference Proceedings of the Novosibirsk October 1970 Switching The Control of Congestion in Packet D W Davies on Symposium of the Second ACM IEEE Proceedings Networks Systems Communication of Data the Optimization Problems in Vol of IEEE Transactions in also 46-49 pp October 1971 COM-20 No 3 June 1972 The Design of a Switching R A Scantlebury and P T Wilkinson Computer Services by Other Access to System to Allow Remote of the Second Proceedings Devices Computers and Terminal the Optimization of Data in ACM IEEE Symposium on Problems also in 160-167 Pp October 1971 Systems Communication 1972 3 June Transactions of IEEE Vol COM-20 No W L rice Controlled Simulation of Packet-Switching Networks Communications ACM Principles on Isarithmic 1973 November Florida Symposium Data Third op 4-49 Physical and P T Wilkinson The National R A Scantlebury ICCC the of Proceedings Laboratory Vata Communication Network 229-238 Conference Stockholm August 1974 pp D W Davies Packet and Future Switching Message Switching Stockholm August Data Communication Networks IFIP 1974 pp 147-150 Congress Isarithmically Controlled L Price Simulation Studies of an Proceedings of the Network Store and Forward Data Communication W IFIP Congress Stockholm August 1974 111 l3 pp 151-154 1 2 Bolt Beranek and Newman Inc 4799 Report No The Events of 1967 and 1968 The years 1967 and 1968 were spent promoting interest in the a IPTO with and government the network the deciding the fundamental structure of contractors writing within both project ARPANET selecting a contractor and work began on the conventions to be used for for request quotation other related 'activities In early 1967 exchanging messages between any pair of computers in and also on consideration of the kinds of communications network and lines data In particular sets to be used for character and block transmission conventions then of the University of Michigan an AA-_h paper on these areas of communication Group was decided include error checking Frank and computer and user identification and retransmission Westervelt it would protocol communication inter-host the that the proposed wrote a position Communication was selected from among the institutions represented and a meeting of the group scheduled The plan considered throughout much of the Michigan was to and data connect communication with technique i e program in all of the computers by dial-up telephone lines so sets each meeting as any calling to other it computer any allow computer up on would 111-14 the computer to establish using a line-switching telephone A small interface to the data set and Report No 4799 Bolt Beranek and Newman Inc phone line and when given a small interface message to computer -another program would perform a message-switching and transmission function deciding how to actually reach computer and transmitting the message to it Washington University in the meeting that refined the Wes Clark is credited with proposing at other then of some point a small computer be inserted between each participant's computer and the further the within phone DARPA line This concept was and the concept of the Interface Message Processor or IMP emerged The IMP checking the would perform retransmission participants' hosts would the functions routing computers of dial-up error and verifications on behalf of which we shall hereafter call Thus the IMPs plus the telephone lines and sets constitute a message-switching network which were to be established formats the between would IMPs define the data The protocols communications The interface between a host and an IMP would be a digital interface of a much simpler sort requiring no host consideration routing It inserting the computer IMPs was of error was clearly IMP was recognized straightforward to network retransmission noted that the major disadvantage the beside each host checking cost of installation of and of another The major advantage of inserting the be design the could fact be that a unified made and implemented without undue consideration of the variations and constraints 111-15 of Report No 4799 Bolt Beranek and Newman Inc the host computer Further an the network evolved it would be much simpler to modify the network of IMPs than to modify all the host computers Finally communications It burdens the IMPs would relieve the hosts of the they was also noticed that if strategic connection were initially necessary IMPs could points within to be implemented faster connected hosts and the network network which entity a network than of be located at the network to concentrate messages over cross-country phone lines likely scheduled to carry a IMPs would be useful in of IMPs was network of directly provided a distinct presenting the network publicly By October 1967 the proposed network was becoming the ARPA Computer topics were under formatting or ARPANET for short discussion message propagation IMP-to-host Network at that protocols queuing dynamic error communication time control It was known A variety of including routing and message message measurements decided as that and 50 Kb communications lines would be used because of the vastly improved response time which could be obtained with these lines as opposed to the previously proposed 2 4 Kb lines be leased eliminating the slow dial-up procedure the telephone tariffs Kb lines The 50 Kb lines were to The nature of available to the government made use of affordable With each IMP normally permanently connected to at least two other IMPs via the leased 50 Kb 111-16 50 lines Report No 4799 Bolt Beranek and Newman Inc the IMPs were to use store-and-forward message handling bits from packets its techniques to provide fast Each IMP was to accept messages of up to 8 000 computer host and to break this Each packet was to be treated independently and on one of the two or more inter-IMP lines at an IMP further it was to be stored At its It destination host expected assembled routed When a packet arrived error checked destination entire message could be was into 1 000-bit and routed on to a packets would be held until an and then delivered to With an average of say three lines per IMP that the it approximately three store-and-forward stages would be necessary to get a message from any one of twenty locations to any other In July 1968 prospective bidders DSSW an RFQ for the network was mailed out to Twelve proposals were received by the Agent Four bidders were rated within the zone of contention to receive the IMP contract and supplementary were requested from each of these bidders technical Final negotiations were carried out with two finalists and one was week before Christmas 1968 briefings chosen in the The contract was awarded and work began the second day of the New Year in 1969 111-17 Report No 1 3 4799 Bolt Beranek and Newman Inc Key Aspects of the RFQ It was specified that evaluated on four criteria in responses to the RFQ would of technical addition to cost 1 Understanding and problems involved 2 Availability of qualified experienced personnel for assignment to software hardware and installation of the system 3 Estimated functional performance and choice of hardware 4 General quality responsiveness commitment to the network concept The RFQ had depth of analysis and corporate provision for a bidders conference and stated that there would be no other opportunity for bidders technical issues with the government four-IMP network A thirteen-month requested to include design IMPs were to take system price operational IMPs and The be installed nine months after start of the months after installation full but to performance period was contract with the contractor supporting them three discuss construction of a prototype IMP implementation and installation of four four to Bidders were asked to provide a system design for a nineteen-IMP network a be responsibility the field for The contractor was required to although portion of the work was a possibility 111-18 in subcontracting a H Report No I 4799 Network Description A Introduction B Functional Description 1 The User Subnet 2 The Communication Subnet C D E F G H II III Bolt Beranek and Newman Inc Functional Description of the IMPS 1 Breaking of Messages into Packets 2 Management of Message Buffers 3 Routing of Messages 4 Generation Analysis and Alteration of Formatted Messages 5 Coordination of Activities with Other IMPS 6 Coordination of Activities with its HOST s 7 Measurement of Network Parameters and Functions 8 Detection and Disposition of Faults 9 IMP Software Separation Protection The HOST-IMP Interfaces The IMP-CARRIER Interfaces Network Performance Characteristics 1 Message Delay 2 Reliability 3 Network Capacity 4 Network Model HOST-HOST Characteristics IMP-Operator Interface Network Contractor Performance Elements of System Design Appendix A ARPA Network Nodes B ARPA Network Topology C IMP Delivery Schedule D Input and Output Facilities for the IMP Operator E ARPA Network Data Rates Between Nodes in Kilobits sec F Data Communications Conventions G Routing Figure 2 Table of Contents from RFQ Statement of Work 111-19 Report No 4799 The IMP Bolt Beranek and Newman Inc specification responsibilities company and clearly among host sites delineated the division of IMP contractor and telephone Each individual host site was responsible for designing implementing for its own convenience software necessary to attach the host to the the hardware network and hardware and software to utilize other hosts on the network telephone company the The was to be responsible for providing necessary circuits data sets and line conditioning equipment utilized the IMP network and The contractor was to by be responsible for providing necessary hardware and software to connect IMPs to each other using the circuits supplied by the telephone company and to connect IMPs to hosts as well as providing hardware and software necessary to implement the procedures which allowed creation of a network of IMPs capable of forwarding messages from one host to another The functional description of an IMP specified the use of 'essages not longer than 8192 bits which packets of not more than 1024 bits would attendant necessity for delays to transmission retransmission take place and permit It error 111-20 in the with the was noted that IMP permit provide queuing space in retransmission into Shorter packets were provision of message and packet buffer space would conversions broken Messages were limited in size to make them manageable for the hosts used to reduce the probability of be event of speed the face of erroneous Report No 4799 transmission Bolt Beranek and Newman Inc A routing algorithm was hypothesized which would take into account the connectivity of the network busyness and message priority forward a packet to the next destination loading periodic information hypothesized IMP updates with and on use a IMP path to the ultimate bajed on exchange of routing and other IMPs and hosts was IMPs were to messages from a local host at the IMP's convenience were also An IMP was to coordinate its activities with other send messages to a local host at IMPs line this information to IMPs and its hosts and perhaps other special hosts take and to be network parameters messages through the host's able at selected and to trace the network but to convenience The times to measure selected the The movements of selected data resulting from these measurements and tracings was to be capable of transmission to host and the measurement activity was to initiation and termination by a host or another were to detect failures examine was from from IMP The various IMP host or reload IMPs from selected network hosts that at each IMP host-specific code to programmers IMP recover capable In particular it was to be possible to thought special and be and thus it the portion a of IMPs and line stop start Finally it site it would be possible for be provided by host site was desirable to protect the rest of the that the host personnel could access and program 111-21 Report No 4799 There were host-to-IMP Bolt Beranek and Newman Inc two particularly interface interesting was to consider multiple hosts per IMP required and 3 of the a standard host interface was to be 1 specified rather than a different one bidder aspects the for cost although each of only host 2 each providing interfaces to one host interface was sufficient program space was to be left to do host-specific character code conversion and repacking of binary messages The interface between required to have hardware to characters calculate real-time clock with fault and status and 20 an iMP and its sense characters check the microseconds information telephone lines was 24-bit network performance CRC Further the control provide and resolution optimized to handle three lines but be capable of Several detect IMP a provide was to be handling six characteristics were specified The average message delay for a short message to go from a source IMP to a destination IMP was to be less than one-half second a fully loaded message errors capacity was network was to considered be The probability of lost messages and very third low in Interestingly and second still network order of importance and was defined to be the maximum bit rate that can node for be input at every have the meSsage delay remain less than one-half a 20 Kb network capacity was hoped for was presented 111-22 A network model Report No 4799 Bolt Beranek and Newman Inc Host-to-host traffic hypothesized that there traffic length flows would were be a estimated trimodal type high rate and short length and it was distribution of medium rate and medium and low rate and long length In addition to considering connected to an IMP and checkout option of multiple hosts the bidder was also asked to consider the provision of memory protection operation the of to new facilitate software simultaneous IMP and to consider what additional hardware and software would be necessary for an IMP to provide a terminal concentration capability for its the network i e no host just terminals 1 - II-2 host or for Report No 1 4 Bolt Beranek and Newman Inc 4799 Chronological Development 1969 to 1975 Within a year after the award of the IMP contract IMPs were installed in the field Hosts were connected to these first IMPs and a series of network measurements Host software an was undertaken was written hosts began to communicate with each other more IMPs and hosts were added became the first operating entity and gradually the ARPANET After six years of development and operation the network was no longer best suited to management an agency development Defense with the charter to sponsor advanced research and and thus network management was transferred Communications Agency by to the In the following subsections we describe the happenings of the years 1969 to 1975 I1I-24 1 4 1 The Groups and the Key People The ARPANET individuals dire tion it Bolt Beranek and Newman Inc 4799 Report No is and development was institutions a joint all Before we describe the ARPANET best to list of effort responsive to development many DARPA's further the principal players and briefly describe their areas of responsibility 11 i'I I III-25 F77-- Report No 1 4 1 1 Bolt Beranek and Newman Inc 4799 Management and Administration of the Network DARPA IPTO Naturally provided encouragement in was the initial the and DECCO The director of IPTO who promoted RML DARPA IPTO played a major role in of the network Taylor DSS-W the at the necessary the development time support helping to fund the network was as Robert well as Lawrence Roberts program manager for the network program Roberts network made certain key architectural led the evaluation team which selected selected other involved contractors the IMP decisions contractor and He also succeeded Taylor as director of IPTO While IPTO set policy for the network who would join the network day by day etc made decisions about IPTO did not run the network BBN provided day-by-day operation and maintenance of the network and BBN and the other contractors out much the of day-by-day business of involved carried the network among themselves without need for daily IPTO supervision Initially IPTO itself filling out took care of ordering telephone lines forms which were sent to DECCO for procurement from and execution by the various initially did its own telephone technical monitoring contractors associated with the ARPANET procurement agency for the contractors 1II-26 companies of Also the DSS-W was used Eventually IPTO various as the attempting to rid Bolt Beranek and Newman Inc 4799 Report No of itself communications the routine providing and lines tedious of aspects ordering monitoring technical of routine dealings the DSS-W functions shifted IPTO contractors Range technical monitoring to the with DECCO and some contractor Force Base in Florida in Laboratory at Patrick Air Measurements RML also had a capability procurement a having addition to technical support capabilitY of the IPTO staff who have were several other members Roberts of the ARPANET besides management the in prominent been There t These have Crocker A included B Dolan Blue C Fields B Wessler S Walker D Carlstrom S and D Russell a strong technical IPTO often played number As mentioned above of the IPTO staff wrote a role in the ARPANET and members of papers describing papers are listed in the ARPANET activity the following bibliography 111-27 Several of these Report No 4799 Bolt Beranek and Newman Inc Bibliography Network T Marill and L G Roberts Toward a Cooperative Time-Shared Computers Proceedings FJCC 1966 p 425-431 of L G Roberts Multiple Computer Networks and Intercomputer Communication ACM Symposium on Operating System Principles October 1967 7p L G Roberts and B D Wessler Computer Network Development to Achieve Resource Sharing Proceedings SJCC 1970 p 5 4 3-5 4 9 L G Roberts A Forward Look L G Roberts 1971 p 4 - 8 ARPA Network Implications Signal August 1971 L G Roberts ARPANET Current Status Spring Conference April 1972 p 7-12 p 77-81 EDUCOM Bulletin Future Plans L G Roberts Network Rationale A 5-Year Proceedings COMPCON 1973 February 1973 p 3- 6 Fall EDUCOM Reevaluation L G Roberts Dynamic Allocation of Satellite Capacity through Packet Reservation Proceedings NCC E 1973 p 711-716 L G Networks Roberts The ARPA N Abramson and F Net in Kuo Eds Computer-Communication Prentice-Hall 1973 S D Crocker J B Postel J F Heafner R M Metcalfe Function-oriented Protocols for the ARPA Computer Network AFIPS Conference Proceedings 40 May 1972 pp 271-279 111-28 Report No 4799 The Network Analysis Corporation 1 4 1 2 Led MAC Bolt Beranek and Newman Inc by Dr Howard Frank of Glen Cove to specify its cost Long Island the Network Analysfs Corporation was put under contract by the topological design of the ARPANET and to analyze performance and reliability characteristics In the process of evaluating any particular network throughput it through the design is necessary proposed of traffic will to the simulate network Then for methods design process for incremental available a delay of a or traffic the design may be altered In this procedure specifying repeated the routes it on so often in the MAC has developed some very efficient changes to a shortest-path algorithm as the network topology is changed discovered flow of flow in the network and the procedure must not be too complex since it must be iterative parameters reliability important to have a facility which the such as cost slightly to improve one of these measures is DARPA routing Further they have faster shortest-path algorithm than was previously taking advantage of the low connectivities usually present in most practical communications networks The following bibliography lists much of the published work by MAC related to the ARPANET 111-29 I Report No 4799 Bolt Beranek and Newman Inc Bibliography NAC First Semiannual Technical Report for the Project Analysis and Optimization of Store-and-Forward Computer Networks June 1970 62p NAC Second Semiannual Technical Report for the Project Analysis and Optimization of Store-and-Forward Computer Networks January 1971 96p NAC Third Semiannual Technical Report for the Project Analysis and Optimization of Store-and-Forward Computer Networks June 1971 109p NAC Fourth Semiannual Technical Report for the Project Analysis and Optimization of Store-and-Forward Computer Networks December 1971 1 2 3p NAC Fifth Semiannual Technical Report for the Project Analysis and Optimization of Store-and-Forward Computer Networks June 1972 92p NAC Final Technical Report for Optimization of Store-and-Forward 1972 120p the Project Analysis and Computer Networks October H Frank I T Frisch and W Chou Topological Considerations in the Design of the ARPA Computer Network Proceedings SJCC 1970 p 581-585 H Frank I T Frisch R Van Slyke and W S Chou Design of Centralized Computer Networks Networks Vol 1 1971 p 43-58 H Frank and W Chou Routing in Vol 1 No 2 1971 p 99-112 H Frank Computer Network Convention Record March 1971 p H Frank Networks Networks England Computer Networks Design -221 2 20 IEEE Optimal 1 no Networks Internation ' and W Chou Throughput in Computer- Communications International State of the Art Report no 6 Computer Infotech 1971 Information Ltd Maidenhead Berkshire p 493-512 R Van Slyke and H Frank Network Reliability I Networks Vol 1 No 3 1971 p 2 79- 2 90 111-30 Analysis Part Report No Bolt Beranek and Newman Inc 4799 R Van Slyke and H Frank Proceedings Networks Conference December 1971 Reliability of Computer-Communication Winter ACM Simulation of the 1971 p 71- 8 2 H Frank and I T Frisch The Design of Large-Scale Proceedings of the IEEE Vol 60 No 1 January 1972 Networks p 6 - 1 1 R E Kahn and L Kleinrock Computer Communication H Frank Practice and Theory with Experience Network Design SJCC 1972 p 255-270 also in Networks Vol 2 No Proceedings 2 1972 p135-166 W Chou and H Frank Routing Strategies for Computer Network Institute Research the Microwave of Proceedings Design Networks and International Symposium on Computer-Communications Teletraffic April 1972 p 301-309 A Kershenbaum and R Van Slyke Computing Minimum Spanning Conference Trees Efficiently Proceedings of the ACM Annual August 1972 p 51 8 -5 2 7 Topological H Frank and W Chou Networks Proceedings of the IEEE Vol 1972 p 1385-1397 Optimization of Computer November 60 No 11 Optimal Design of Computer Networks in Computer H Frank Networks R Rustin Ed Prentice-Hall 1972 p 1 6 7-1 8 3 in Routing Policies Deterministic and Adaptive M Gerla Packet-Switched Computer Networks ACM Third Data Communications Symposium Florida November 1973 p 23-28 Slyke Simulation of Centralized R 'Van H Frank W Chou Computer Communications Systems ACM Third Data Communications Symposium Florida November 1973 p 121-130 W Chou and A Kershenbaum A Unified Algorithm for Designing Multidrop Teleprocessing Networks ACM Third Data Communications Symposium Florida November 1973 p 148-156 H Frank Providing ACM Third Components November 1973 p 161-164 Unreliable Reliable Networks with Data Communications Symposium Florida I Gitman R M Van Slyke and H Frank On Splitting Random Channels Proceedings of the Accessed Broadcast Communication on System Sciences Conference International Seventh Hawaii January 1974 5p 111-31 1 4 1 3 Bolt Beranek and Newman Inc 4799 Report No The Telephone Companies Building a nationwide communications network means doing business with a number of telephone companies network deal is to have overseas or foreign components with Telephone further various international Telegraphs This carriers task was and if the one must also national Post handled by DECCO which negotiated with the relevant telephone companies and obtained the specified service at the best price from UCLA to RAND for example procured from General Telephone In the case of a circuit most likely the service would be the dominant telephone company in the Los Angeles area in the example just given a requested circuit fell completely within the jurisdiction of a single telephone company To handle all instances when the requested service jurisdictions of more than one telephone company utilized their Long Lines division the ARPANET circuits service was Long and in necessary to make up the which in requested the the Bell System the case of many of the company from which DECCO Lines spanned procured the turn procured the components service from the regional telephone companies For the first several years of the ARPANET development Long Lines customer representatives Fraser Bill Gordon and 111-32 to DARPA were Ken Stanley in A turn the Al customer Report No 4799 Bolt Beranek and Newman Inc representative's job is to make the customer aware of of service the and to keep him happy with the service he available receives Fortunately understood that for the ARPANET the Bell requirements rather than having to rely grasp on the miscellaneous dealings with local telephone companies Long Lines System a customer building a nationwide network needed the assistance of some central individual with a broad the kinds representative of usual Thus the who from his position in Long Lines was already in contact with his counterparts in the many regional telephone companies was permitted to use his network of contacts of coordination informal to provide the System Bell entire service to the ARPANET For instance installing a telephone circuit between two IMPs requires that the IMPs and the telephone companies ends all be ready simultaneously date date if Similarly it strain to make a both It is useless for the IMP supplier and one of the telephone companies to strain to scheduled at make a the other telephone company cannot make that is useless for the date if telephone companies the IMP supplier cannot to The Bell System customer representattve took it upon himself to coordinate all such interdependent events willingness to By virtue of the Bell provide this critical coordination an extremely smooth and efficient relationship has been built up IMP suppliers DARPA and the member System's between companies the of the Bell 111-33 I _ ' i i o J i l I I' I i Report No System Bolt Beranek and Newman Inc 4799 For a network of the size and complexity of the ARPANET there has been surprisingly little trouble with and operation of the telephone circuits 1 A the procurement Report No 4799 1 4 1 4 Bolt Beranek and Newman Inc Bolt Beranek and Newman Inc The contract to construct awarded to Bolt Beranek and the Newman IMP Inc for the ARPANET was BBN of Cambridge Massachusetts where leadership of Mr installed BBN continued to play a central role in the evolution of the network the it was Frank Heart operating it development carried Once out in a group under the the and maintaining it first IMPs were as well as doing necessary for several major enhancements of its capability In addition to Frank Heart a number of other individuals at The BBN have been involved with the development of the ARPANET names of many of these individuals may be found as authors of the papers on the ARPANET which have come out of BBN Robert individual several years Kahn deserves particular However mention has After as a principal member of the group working on the network at BBN he moved to the IPTO office at DARPA he one probably from which done more to promote and support the continued advance of packet-switching technology than any other individual A bibliography of ARPANET-related reports and papers wri ten L by members of the BBN staff is included as Appendix A I - -I 111-35 Report No 1 4 1 5 Bolt Beranek and Newman Inc 4799 The Network Information Center The accessibility of distributed resources carries the need for an information service either distributed that enables users to learn about A contract was awarded to for the it centralized or those resources the Stanford Research Institute to develop and operate a Network Information Center established with ARPANET implementation of the network in With 1969 the NIC to beginning be of construction also began on the NIC at SRI The NIC provided several services network participants maintained a list and distribution lists interest groups within the various It network for various special community document series was maintained of An archive Documents could be sent to the NIC with instructions for duplication and distribution the membership or one or more of the ipecial highly structured display system data base called NLS over the network A list hosts the throughout of interest groups construction manipulation to A and was made available on-line for use was kept of the resources available network The various on ARPANET protocol specifications were maintained on-line at the NIC The NIC has had a hand in the proauction or distribution hundreds and hundreds of documents related to the ARPANET of these documents describe the activities of the NIC itself 111-36 of A few or Report No are 4799 otherwise Bolt Beranek and Newman Inc of special interest bibliography of these follows I I 1 II1-37 within the ARPANET a Report No 4799 Bolt Beranek and Newman Inc Bibliography D C Engelbart Network Information Augmented Team Interaction Stanford Augmentation Research Center January 30 Center and Computer Research Institute 1971 105p D C Engelbart On-line Team Environment Stanford Research Institute Augmentation Research Center June 2 1972 272p D C Engelbart Coordinated Information Service for a Discipline or Mission-oriented Community Time-sharing Past Present and Future Second Annual Computer Communications Conference 1973 pp 2 1-2 4 Current Catalog of the NIC Collection NIC No Current Directory of Network Participants Network Resource Handbook NIC No 6740 Current Network Protocols NIC No 7104 111-38 5145 NIC No 5150 Report No 1 4 1 6 4799 Bolt Beranek and Newman Inc The Network Measurement Center One of the influential early studies of communications networks was performed by Leonard Kleinrock and his 1962 Ph D thesis Message Flow and Delay York in 1964 put Communication Nets by Dover Kleinrock of algorithms in Stochastic Publications New network can be made very low and considered communications several networks aspects and of the their underlying developing a precise model for networks Kleinrock today a faculty member at UCLA UCLA was selected installation which was to to Measurement allow be used Center to be the site well as NMC the first to NMC support The the tasks of a each Network NMC had the responsibility ARPANET for performance While Kleinrock himself was the guiding force over the years he had a series of students or staff members who supervised the day-to-day measurement work Gerald Cole IMP direct measurements based on statistics gathered by the IMP program the of early connection of an SDS SIGMA 7 host much of the analysis and simulation of the at in rest one of the early and persistent objections to networking as reported This study proved convincingly that message delays to operation MIT reprinted a large store-and-forward thus is at is Vinton Cerf obtained Holger Opderbeck a UCLA Ph D and including William Naylor although not necessarily for their NMC work 111-39 Report No 4799 Bolt Beranek and Newman Inc There has been a series of doctoral di-ssertations written by students at the UCLA School of Engineering work of the NMC A list and of of these theses related to the as well as other relevant publications by the faculty and students associated with the NMC is included in the following bibliography 111-140 Report Nc 4799 Bolt Beranek and Newman Inc Bibliography L Kleinrock Communication Nets Stochastic Message Delay Dover Publications New York 1964 2 0 9p Flow and L Kleinrock Models for Computer Networks Proceedings of the International Communications Conference June 1969 p 21 9-21 16 L Kleinrock Methods for Computer Network Modela Comparison Proceedings ofof Solution the Computers and Communications Conference October 1969 L Kleinroek Network Design Analytic and Simulation Proceedings SJCC 1970 p 5 Methods 6 in Computer 9-579 L Kleinrock and G L Fultz Adaptive Routing Techniques for Store-and-Forward Computer Communication Networks International State of the Art Report No 6 Computer Networks Infotech Information Ltd Maidenhead Berkshire Efigland 1971 p 541-562 also in Proceedings of the International Conference on Communications Montreal Quebec Canada June 1971 p 39 1-39 8 G Cole Performance Keaourements on the ARPA Computer Network Proceedings of the Second ACM IEEE Symposium on Problems in the Optimization of Data Communications Systems October 1971 p 39-45 also in IEEE Trans on Communications Vol COM-21 No 3 June 1972 p630-636 V G Cerf and W E Naylor Storage Considerations in Store-and-Forward Message Switching Paper presentec at WESCON Conference September 1972 8 p V G Cerf and W E Naylor Selected ARPA Network Measurement Experiments Proceedings COKPCON 1972 September 1972 L Kleinrock Computer Networks in Computer Science A F Cardenas et al Eds John Wiley and Sons Inc New York 1972 p 241-284 L Kleinrock Performance Models and Measurement of the ARPA Computer Network Proceedings of the International Symposium on the Design and Application of Interactive Computer Systems Brunel University Uxbridge England May 1972 30p L Kleinrock Survey of Analytical Methods Networks in Computer Networks R Rustin Ed 1972 p 1 8 3-205 111-41 in Queueing Prentice-Hall Bolt Beranek and Newilnan Inc Report No 4799 D Cantor and M Gerla The Optimal Routing of Messages in a Computer Network Via Mathematical Programming Proceedings COMPCON 72 September 1972 p 167-170 L Fratta M Gerla and L Kleinrock The Flow Deviation Method An Design Networks Approach to Store-and-Forward Communication Network Vol 3 No D U Cantor and M Gerla 2 1973 Capacity p 97-13 4 Allocation in Distributed Computer Networks Proceedings of the Seventh Hawaii International Conference on System Sciences January 1974 p 115-117 G Cole Computer Networks Measurements Techniques and Experiments PhD Thesis UCLA-ENG-7165 Computer Science Department School of Engineering and Applied Science UCLA October 1971 350p G L Fultz Adaptive Routing Techniques for Message Switching Computer-Communications Networks PhD Thesis UCLA-ENG-7252 Computer Science Department School of Engineering and Applied Science UCLA July 1972 4 18 p M Gerla The Design of Store-and-Forward S F Networks for Computer Communications PhD Thesis UCLA-ENG-7319 Computer Science Department School of Engineering and Applied Science UCLA January 1973 30 0 p Kleinrock L Queueing Systems Interscience New York 1975 Volume I Theory Kleinrock L Queueing Systems Volume Applications Wiley Interscience New York 1976 II Kleinrock in L Scheduling Queueing and Delays Wiley Computer Time-Shared Systems and Computer Networks in Computer-Communication Networks Abramson N and Kuo F eds Prentice Hall Englewood Cliffs New Jersey 1973 pp 95-141 Kamoun F Communication Science Design Networks Department UCLA April 1976 Considerations PhD School Thesis Large Computer UCLA-ENG-7642 for Computer of Engineering and Applied Science 407p Ziegler J and L Kleinrock Nodal Blocking in Large Networks ICC Conference Record Montreal Canada June 1971 pp 39-9 to 39-15 11-i Report No 14799 Bolt Beranek and Newman ILnc Kahn and L Kleinrock Computer Communication R H Frank Network Design--Experience with Theory and Practice AFIPS Conference Proceedings Vol 40 SJCC Atlantic City New Jersey May 1972 pp 255-270 Also in Networks Vol 2 No 2 1972 pp 135-166 Kleinrock L Analytical Techniques for Computer- Communication Networks Computers and Communications Proc of the joint IBM-University of Newcastle upon Tyne Seminar University of Newcastle upon Tyne England September 1973 pp 63-90 Kleinrock L and W Naylor On Measured Behavior of the ARPA Network AFIPS Conference Proceedings Vol 43 NCC Chicago May 1974 pp 767-780 Kleinrock L Resource Allocation in Computer Systems and Computer Communication Networks Proc of the International Federation for Information Processing Stockholm Sweden August 1974 pp 11-18 Opderbeck H and L Kleinrock The Influence of Control Procedures on the Performance of Packet-Switched Networks National Telecommunications Conference Record San Diego California December 1974 pp 810-817 Naylor W E A Loop-Free Adaptive Routing Algorithm for Packet Switched Networks Proc of the 4th Data Communications Symposium Quebec City Canada October 1975 pp 7 9 to 7 14 Kleinrock L W Naylor and H Opderbeck A Study of Line Overhead in the ARPANET Proc of the IIASA Conf on Computer Communication Networks October 1974 pp 87-109 Also in Communications of the ACM Vol 19 January 1976 pp 3-12 Kleinrock L and H Opderbeck Throughput in the ARPANET-Protocols and Measurement 4th Datd Communications Symposium Quebec City Canada October 1975 pp 6-1 to 6-11 also in IEEE Trans on Communications Vol COM-25 January 1977 pp 95-104 Kamoun F and L Kleinrock Analysis of Shared Storage in a Computer Network Environment Proc of the 8th Hawaii International Conf on System Science Honolulu January 1976 pp 89-92 Kleinrock June 1976 L pp ARPANET Lessons 20-1 to 20-6 Proc 111-43 of the ICC Philadelphia Report No Bolt Beranek and Newman Inc 4799 Kleinrock L and F Kamoun Packet-Switching Networks Congress Sydney Australia Data Communications Through Large Proc of International Teletraffic November 1976 pp 521-1 to 521-10 W and L Kleinrock On the Effect of Periodic Routing Naylor Dallas Updates in Packet-Switched Networks Proc of the NTC Texas November 1976 pp 16 2-1 to 16 2-7 Kleinrock Computers 1326-1335 IEEE Trans L On Communications and Networks 25th Anniversary Issue Vol C-25 December 197 6 on pp On The Topological Design of Gerla M and L Kleinrock IEEE Trans on Communications Distributed Computer Networks Vol COM-25 January 1977 pp 48-60 Kamoun Hierarchical Routing for Large Kleinrock L and F Computer Performance Evaluation and Optimization Networks Networks Vol 1 No 3 January 1977 pp 155-174 Kleinrock L Performance of Distributed Multi-Access ComputerCommunication Systems Proc of IFIP Congress Toronto Canada August 1977 pp 547-552 Analysis of Buffer Allocation Kermani P and L Kleinrock 2 Vol Schemes in a Multiplexing Node ICC Conference Record Chicago Illinois June 1977 pp 30 4-269 to 30 4-275 I 111-44 i ' ' o II l '171 iiInowiI Report No 1 4 1 7 4799 Bolt Beranek and Newman Inc The Network Working Group and Host Involvement The initial design of the ARPANET as contained in went some way toward specifying communications and for attention was given being left To host-to-host and J formats for AP-to-host communication to host-to-host the Less explicit communication hosts with a little problems of SRI DARPA assigned After an initial Rulifson met again in this area the impetus to work on the the meeting summer problem S and Crocker fall of continue discussion of host-to-host protocol issues thinking inter-IMP for host sites to work out among themselves provide Shapiro certain the RFQ was at a very high level e g to Elmer S Carr 1968 to Their early the feasibility of creating a portable front-end package which could be written once and moved to all network hosts a host desiring to send another host would first data to send a data description to the receiving host which instructed the front-end package at the receiving host to hcw interpret Valentine's Day and 1969 representatives contractor data coming the first from the sending meeting of host the was held at BEN in from NMC and NAC the middle of host On representatives along 4ith the IMP an enormous snow storm In April 1969 Comments RFCs a series of working notes called Request for was established which could be circulated to let III-45 Report No 4799 Bolt Beranek and Newman Inc others know what they were doing and to obtain the reactions involvement of other interested parties and They called themselves the Network Working Group NWG The NWG eventually grew quite from almost every host large with site in the network participating mountains of paper was circulated describing various protocols There From about the time it until near the chairman of the NWG too large but he much meetings committee which provide protocol working of left DARPA he would go to DARPA Stephen Crocker served as its work was done -over on also occasional mass meetings the network iet e g to for discuss general grown large numbers of From Crocker and protocol to this point those issues of and various the group interested in the Remote Job Entry Even after the big meetings stopped notes commenting and were limited to those of an executive protocol guidance subcommittees and By the beginning of 1972 the NWG had hosts were communicating onward were was decided that time representatives most participant were circulated to most other participants in the network Gradually the activities of the NWG began to diminish of the host site personnel who had originally been active on to 0se the defined protocols rather than becoming involved in other tasks Many moved and new users joining the network tended to their ri 1- 111-46 - k' r 4 - Report No 4799 specification Bolt Beranek and Newman Inc As Crocker's time for the NWG group became increasingly limited he appointed Alex McKenzie and to serve jointly in his place after process Postel McKenzie and Postel interpreted their task to be one of codification and coordination and Jon primarily a fe4 more spurts of activity the protocol definition settled for the most part inio the status of a maintenance effort Further activities of the Network Working Group will be described in the section below on the development of host-to-host protocol 1 III-i7 Report No 1 4 2 4799 Bolt Beranek and Newman Inc Initial Subnet Design BBN's proposal for the IMP was for the most the with however requirements the BBN IMP design diverged from those The BBN compliant In some important respects of tne RFQ added constraints that were not in part requirements the RFQ design took the idea of inserting a communications processor between the host and the network to a logical it specified functions that the IMP be used only binary Thus conversion for a host of the separation between the design did not provide for IMP repacking messages programming extreme for communications and that there 1 e maximum logical IMP and host of or on behalf of a host or for doing character Further IMP and the design precluded all host any control of the IMP by regular hosts The initial subnet design messages between an IMP and its in the RFP also specified hosts minimal control many fewer than envisioned As the network later developed it proved useful to add additional su'ch control messages The IMP design called for a hardened robust statistics in the face of physical IMP which would and electrical abuse have shown that hardening did help it was While eventually decided not be be worth the added price of the hardware 111-48 be Report No 4799 The Bolt Beranek and Newman Inc capability design This has to loop all interfaces was included in proved to' be enormous of the operational importance A tape decision was made to reload IMPs initially and each IMP was provided with a paper tape reader always intended that this procedure replaced the paper tape Program local would by loading through the network reader simplification in debugging that at each IMP sooner and it was or was It was later be The use of probably a useful the beginning terminal knowledge in was the also specified to take place from a interests of simplicity with tae cross-network debugging could be added later cross-network debugging capability first from paper was added even before A the IMP was delivered A delivery schedule of one IMP per month in months eight through eleven after contract start instead of all four at month nine was assumed In keeping w1th the strict IMP was to periodically gather to statistics independence of host and IMP routinely and transmit the them specified hosts rather than permitting hosts to control the IMPs' statistics-taking satisfactory 111-419 capabilities This proved Report No The 4799 Bolt Beranek and Newman Inc initial IMP design was responsive to the RFQ in one particular way which was changed at BBN DARPA host per the and host representatives IMP and clear connect to delivered that a many single the an option host sites IMP Thus meeting although it Almost it had more than one host to before four also asked immediately the first design was changed to permit two hosts on an IMP along with five between The RFQ had called for one six circuits per IMP that more hosts per IMP be became first IMP three or three lines was or four This was a simple change to implement In the area of the host-to-IMP protocol design specified this protocol as required aspects of the initial IMP Unfortunately some the host-to-IMP protocol had significant detrimental effect on the design and performance of the other prutocols Particularly unfortunate have been the acknowledgment system retransmission system and the message the identification system initially suggested by the host-to-IMP protocol It is crucial for the IMPs to limit the rate of flow of host traffic into the net to the rate at which that traffic is taken out The IMPs' first attempt suggesting in order to prevent subnetwork congestion which was insufficient that each message in the sending host until an took the form being of a conversation should be held by end-to-end 111-50 acknowledgment for the i Report No previous 4799 Bolt Beranek and Newman Inc message was received This suggestion was adopted as part of host-to-host protocol upon which all the higher protocols are based host-to-host ARPANET As a consequence protocol response connection is time destination-to-source using 50 the bandwidth of a single severely Kbs acknowledgment standard limited given the lines waiting for I a between messages typically limits connection bandwidth to about 10 Kbs in contrast to the 40 Kbs possible with a constant stream of messages It was originally thought that the ARPANET would lose a message so seldom that there was no point in with message retransmission hosts ever bothering Unfortunately resolving various possible lockups has required the subnetwork to discard a message occasionally and the topology of the network has evolved into long series of machines and lines that increase of involuntary message loss followed the initial retransmission message loss in which is However thought and Given the did the the host-to-host protocol not realities of provide the sensitive to any As robust although it part of the message probability of protocol abnormality host-to-host protocol and protopols based on it for the network and given the host-to-host inordinately particularly probability has not the proved has been reliable host-to-IMP acknowledgment system described Pbove the protocol host-to-IMP end-to-end protocol 111-51 MW go ' i i tt tm li I iili ili I Report No 4799 specified an Bolt Beranek and Newman Inc 8-bit that all messages identification message identification number and suggested in a single number in conversation fact carry messages this with same different identification numbers were not guaranteed to be delivered in the order sent uniquely Eight bits for is probably insufficient sent identify the purpose of possibly required retransmissions outstanding messages when successive messages in are to a conversation without waiting for an end-to-end acknowledgment Use of the small 8-bit message identifier was one of the factors that prevented reliable high-bandwidth connections The IMP host protocol has been changed longer necessary message order considerations to is cases wait for the now preserved requiring unambiguou-ly reported to the so that is no end-to-end acknowledgment except message sending host for priority retransmission and identifier has been expanded to a sufficient size 111-52 it the are message Report No 1 4 3 4799 Bolt Beranek and Newman Inc Subnet Development The first four IMPs were developed and installed on schedule by it the end of 1969 No sooner were these IMPs in became clear that some provision was needed to relatively distant from an IMP i e the expected 50 feet interface drivers was was Augmented distant simply by these distant interfaces made clear that needed on the host IMP interface assumed there would be no errors on connect hosts up to 2000 feet instead of Thus in early 1970 a developed the field than heftier error Previously such a IMP host local it line control had been hard wired connection By mid-year of 1970 were being carried out quickly corrected Also by a series of network performance tests These uncovered some flaws which were and some problems which looked more worrisome mid-year a rudimentary version of the network control center was established at BBN As the year wore on network the IMP development continued between two IMPs sites continued program continued the first 230 4 and Kb to be added to be improved circuit was to the NCC tested design for a version of the IMP able to support direct terminal connection was called the Terminal IMP TIP 111-53 begun The latter was Report No 4799 Bolt Beranek and Newman Inc In 1970 major problems with the IMP flow control and storage allocation techniques were demonstrated even after these problems were It is interesting that demonstrated and they were serious enough to completely halt network operation under certain circumstances the for many months many implemented in network continued to give adequate service while improvements were designed and The hosts were simply asked to not use the network the way that caused the subnetwork problems and the hosts did as they were asked About three-quarters T1Ps were delivered of the way through 1971 the first two providing ARPANET access for the first time to users without their own hosts or access to terminals on soine other organization's host By the beginning of 1972 it 'as recognized that even the distant version of the IMP host interface was not sufficient design for a circuits worth a was little asynchronous IMP host begun interface for additional comment the network the hosts for RFQ may have been of some benefit in its communications initial bit-serial IMP host in an is interface effort to was simplify This non-standard interface simplifying the host connection greatest virtue was the separation 111-54 l The non-error-cuntrolled connection over The evolution of the IMP host interface essentially specified in However use and it put between i '1 Report No the IMP 4799 and Bolt Beranek and Newman Inc the host The IMP and host did not have to worry about each other's word size and they about constraints each other's timing having to worry about these issues operation However hodge-podge of distant first this interface operation than was very elegant did not It would variations its e g to HDLC use an it would used be standard communications interface Also between the expanded While this option two network sites a massive change in IMP it was never software was to correct the previously discovered flow control and storage allocation problems new version of increments more clearly half of 1972 the TIP's capability was successfully very elegant f for a which need not fear to support TIP-to-TIP magnetic tape transfers undertaken in for every IMP host connection In the first was network and none except the For any new network industry worry resulted designed predecessors the now proven packet-switching technology better delayed also each to seems likely that have interface have the IMP In the second half of the year program was and the design of a new released in many the small ten times more powerful IMP was begun The beginning of 1973 brought the first the network traffic SIII1-55 from rapidly California increasing to a satellite Hawaii number of Also subnet link in with network reliability Report No 4799 problems Bolt Beranek and Newman Inc developed which had to be corrected pair of TIPs had been shipped to Europe London first These time circuits brought for use By mid-year in Norway numerous operational problems circuits had to be obtained from a foreign were relatively slow at 9 6 Kb a and For the PTT and like Hawaii the these TIPs were on a long spur off the network rather than being doubly connected as IMPs typically are be delivered During 1973 nodes continued to but there began to be a low level of node locations to optimize the use of various TY ' and came as sites on and went off the switching configurations network improvements were also made to correct problems with the algorithm As 1973 ended the first of Certain routing very distant hosts were connected to the network over telephone lines In 1974 there were major efforts to make operationally usable control and network more Subnetwork reliability was improved as was TIP-to-host communication access the reliability accounting and subnetworks of hosts were developed Methods for providing TIP pa t tioning Methods were of logical developed to selectively reload sections of IMP memory In 1975 network development slowed up and the network took on more and more of an developments in modification of operational appearance 1975 included delivery of the first the IMP Major network Pluribus IMP and TIP software to support more than 111-56 Report No 4799 Bolt Beranek and Newman Inc sixty-three IMPs in the network and attachment of the Satellite IMPs to the network first two By the end of 1975 the network was under DCA management Looking back I appears the subnet development between 1969 relatively smooth and although there were many times during that period when those intimately involved felt they were to solve one crisis or another and the basic technology I robust enough that 1975 and trying The network grew slowly enough implementation many problems was flexible and both major and minor which naturally cropped up with this new development were for the most part corrected before they obstructed the work of too mauy users The fact that the network was also part of an experiment no doubt also made users more tolerant I III-57 tita tttll'i ll NNWllII11 Report No 1 4 4 4799 Bolt Beranek and Newman Inc Host Protocol Development Specifications generally called the IMP-to-host protocol exist for the physical and logical host and its however to processes IMP message This protocol is specify running in the methods transfer between a not sufficient by itself of communication two possibly dissimilar hosts between Rather the processes must have some agreement as to the method of initiating communication forth the interpretation of Altnough reached by each communication to minimize it pair a the network-wide to of hosts or processes more general arrangement is amount of communication facilitate data and an additional specifications Working order necessary for the host organizations Group exchange of communicartions protocols the services of or ideas NWG introduced and to formulate for host-to-host communications wherein the higher layers lower l3yers the of report As shown IMP-to-host protocol layer in the in Figure The next layer figure specifies 111-58 3 advantages the lowest layer is called methods the of of protocol disadvantages of the layered approach are discussed elsewhere thiz in desirable in implementation Accordingly interested The NWG adopted a layered approach to the specification use so would be possible for such agreements to be formed a group the Network above transmitted and in the host-to-host establishing Report No Bolt Beranek and Newman Inc 4799 iA ad hoe S Host Host FTP Figure 3 -- Host IMP Layered Relationship of the ARPANET Protocols communications paths between hosts managing buffer space at each end of a communications path Protocol or etc Next the Initial specifies a standard way for a remote user or ICP process to attract the attention of a network host to that using Connection The host preparatory ICP provides the analog of the user pressing the attention button at a local terminal on a host the next protocol hosts is layer the Telecommunications Network which was designed to support terminal TELNET is In or TELNET access to remote a specification for a network standard terminal and the protocol for communicating between this standard terminal and a host The next logical protocol layer consists of function oriented protocols two of which File 111-59 Transfer Protocol FTP Report No 4799 and Remote Finally Job Bolt Beranek and Newman Inc Entry at any point in protocol RJE are shown in the layering process it is the figure possible to superimpose ad hoc protocols In events in the following subsections we discuss in some detail the the evolution of the host-to-host and TELNET protocols and the events in the evolution of a number of other protocols in somewhat less detail 111-60 oUm o oo ooo -'o'cl ' lo i- 'o oool 'o om't _ _ oo_otao _ oo--- Bolt Beranek and Newman Inc 4799 Report No 4J 1 4 4 1 Host-to-Host Protocol The Network Working Group was established in December 1969 an initial early 1969 host-to-host protocol had been specified which supported communication between a terminal on one host a process on another December 1969 the initial to host At a meeting in protocol specification plan would not support transmission over the network back and get it By the of was described because the electronic mail He instructed the Network Working Group to go right spring of 1970 several successive versions of a host-to-host protocol had been developed and a relatively formal meeting of the NWG was held at UCLA before mid-year at which version latest and Salt Lake City in Lawrence Roberts of DARPA who was unhappy with it initial By of the protocol was described described protocol were very negative the Reactions to the In June of 1970 there was a series of meetings held at UCLA and Harvard at which from finally to settle upon a these two institutions tried host-to-host protocol and specify how it should be people implemented In August of 1970 some of the more general and some thought more exotic aspects of the host-to-host protocol being considered were ordered dropped from the protocol by Barry Wessler of DARPA thus administratively clearing away some of had prevented agreement Tha those issues which NWG discussion continued at the 111-61 Report No 4799 Bolt Beranek and Newman Inc 1970 Spring Joint Computer Conference in particular there was discussion between Crocker and Roberts regarding the formality to be and DARPA approvals required sought for the protocol forth Another NWG meeting was held at the Fall Conference in November 1970 in At a NWG meeting held in of Illinois host-to-host desirable a to necessary wrote an interim report It see what This appears 1971 appointed changes at the University to look were design of where it days met for two and then reconvened a month later in that with Los the efforts of this committee committee protocol was finally host-to-host ARPANET the the immediately known as the host-to-host protocol glitch cleaning the at subcommittee went directly from Massachusetts Illinois to Cambridge Angeles was Computer Texas mid-February subcommittee protocol or Houston Joint and so coming close to being settled At about this same time DARPA was beginning to pressure not also to get it only McKenzie implemented by the hosts took on the At a NWG meeting at the Atlantic City task of in writing specification of the host-to-host protocol -protocol great to get the host-to-host protocol settled but Spring Joint Computer Conference in Alex exert not to a May 1971 definitive invent new but to write down what had been decided 111-62 il i Report No 4799 In M I T Bolt Beranek and Newman Inc October 1971 the and was preceded by final a programmers' workshop at which January' a 1972 McKenzie ARPANET document In describing the protocol was host-to-host II1-63 t meeting was held at implementations were clarified and eliminated essentially unchanged since A NWG differences in published and the 1 big protocol has remained Report No 1 4 4 2 4799 Bolt Beranek and Newman Inc The Evolution of TELNET Early in the development of the ARPANET it became clear that a major function of the network would be to provide remote use of interactive systems his local host if To allow a user at a terminal connected to to control and use a process in he were a local user of that remote host was required example The problems to be the typical host expects its physically attached to the a remote host as a special mechanism overcome are legion for interactive terminals to be individual ports of its hardware terminal scanner rather than logically attached via a multiplexed connection to the network only with half-duplex 134 5 baud terminals with line-at-a-time certain to communicate characteristics physical echo characteristics character-at-a-time e g no character echo e g EBCDIC character set while a remote user's terminal might have different baud a given host expects completely full-duplex ASCII character set 300 The TELNET protocol was an Pttempt to provide the special mechanism necessary to permit such communication As early as 1969 a few hosts had been programmed on an Ad hoc basis to permit terminal access from another host In 1971 an NWG subcommittee was formed to consider the general problem of supporting interactive use arbitrary remote terminals of arbitrary hosts by There was great controversy 111-64 users at in the Report No 4799 committee Bolt Beranek and Newman Inc discussions focusing on four issues connection establishment echoing late enough 1972 there was character set By and interrupt capability consensus implementation of an early version of so the that TELNET widespread protocol had been accomplished Despite protocol declare widespread implementation its heavy and effective use it complete discussion of the early and numerous of it TELNET attempts continued to There were several problems with the early version 1 Despite the attempt to permit a minimal well suited to the constraints of small hosts was no well-defined minimal some TELNET feature implementation it was had to implementation not be desired provided other implementation commanded its 2 The control structure unless made some it was there Even for in a if given case some use inadequate For example exceedingly constraining assumptions were was possible for the connection implementation to loop two commanding ends each of a other TELNET to take opposite actions 3 The asymmetry of TELNET connections precluded from initiating certain functions such one end as echoing 111-65 S o o nl n w i lW I Hi iW mln i ilill lrl Report No 4799 Bolt Beranek and Newman inc behavior This seriously constrained the use of TELNET processes protocol for character communication between not serving a terminals role for which otherwise have been well suited and for already frequently used would it which it was in the absence of any better protocol 4 The issue of interfacing character-at-a-time hosts to line-at-a-time hosts was poorly handled By early 1973 it had become apparent that minor adjustments to the early TELNET protocol would not solve these that some fundamental changes were needed met and with the previous experience to several fundamental principles problems and A new subcommittee guide them developed These new principles when added to the earlier principles of the Network Virtual Terminal and the remote interrupt synch protocol which solved most mechanism of resulted in a revised TELNET earlier the problems that had precluded universal acceptakice of the protocol There was such enthusiasm for the schedule for rapid within the year out However the implementation reasons was that a laid the new TELNET protocol Only in the past year implementations been widely available several version implementation of proceeded more slowly than expected new have In retrospect there were for the delay in the implementation 1 at the 111-66 V _ _ z ro' - ot -ioillb 'ti i oo _ o _ o _ I o - i -i iii o- -- - o _ _ Report No 4799 time r-vised the Bolt Beranek and Newman Inc protocol implementation was scheduled implementation of the initial version had been completed and host system managers 2 implementation the network had not of protocol existed none most was found and by most in clean implementors 4 memory TELNET several years and chose to provide a complete implementation of in implementation parallel for with the initial the most prevalent user and time-consuming protocol and it has been the revised protocol accepted standard for is widely implemented and used III-67 TIP's thus implicitly relieving pressure on the server hosts to implement the new second phasing over from the initial protocol to the revised the TIP proved to be very difficult because of the limited The a 3 despite initial belief that a the revised protocol to operate host for was declining and the network was entering a period consequently protocol resources about this time DARPA's research interest of status quo operation method budgeted Report No 1 4 4 3 Bolt Beranek and Newman Inc 4799 The Evolution of the Other Host Protocols There are several other host protocols the evolution of which should be briefly mentioned The Data File Transfer Protocol started out as two protocols Transfer Protocol over-simplify the Data and a File Transfer Transfer Protocol to specify how it was transferred control portion After relatively little little final data push to additional work was done effort and a good the to bit Transfer Protocol Eventually Transfer Protocol alone was defined with a To was to specify the format of data being transferred and the File was Protocol a the File portion and a specify the FTP consisting only of a clean up fundamental aspects of the protocol of work reconciling the reply codes that different hosts used to indicate FTP-related events Before a Remote Job Entry protocol could be defined by the NWG as a whole UCLA's IBM 360 91 host a batch oriented host needed some RJE-like protocol with which to serve a few users who wanted host early Thus access to the computing power of that particular led by the UCLA group a protocol called the Job Service or RJS protocol was defined and implemented eventually got such a The NWG around to working on the problem of a Remote Job Entry protocol and define Remote undertook protocol a relatively However 1-68 massive effort to by the time the RJE protocol Report No 4799 Bolt Beranek and Newman Inc definition was finished implemented to interim RJS protocol of the hosts on batch RJE there the network was little protocol specified half a doz i or Thus but so hosts had already Since these included most interested in supporting remote incentive for them to implement the new today the to our knowledge is RJE protocol is carefully not implemented anywhere and the RJS protocol prevails Moving upward in the subject of sophistication another protocol that early discussion was one for graphics was Several versions of a graphics protocol were specified but until recently there was Recently never as widespread part of the implementation ACCAT of experiment any an of them operational graphics protocol has been developed In addition to the host-to-host protocol which specified after much iteration were suggested by various each suggested members of the NWG an alternative protocol of derived suggestion More a recently protocol as National Software works project part of the Robert Thomas David Even after the there was some with Before the Richard Kaline and adoption of the host-to-host protocol experiments finally a number of alternative protocols host-to-host protocol was settled upon Walden was from discussion the Walden DARPA-sponsored Stuart Schaffner and their colleagues have designed and implemented a host-to-host 111-69 Report No Bolt Beranek and Newman Inc 4799 known as TCP deserves Near the time of the formation of the International Network protocol Another protocol known 3s MSG special mention Working interest Group as network interconnection discussions began on 3 -t idard inter-network protocol particularly one which ootld the began to be of great ARPANET host-to-host protocol ore of the shortcomings zrret of A- the AFIPS 1973 NCC in New ideas York City a meeting was held at which certain host-to-host correspondence Stanford discussed After Robert Kahn of DARPA and Vinton were protocol got together Other members of INWG protocol a new some additional Cerf then of a protocol known as TCP perhaps not satisfied that TCP represented continued an international standard host-to-host designed and for also Cerf still developing in participated effort TCP quickly became DARPA's choice of protocol to be used in the another this later host-to-host situations where the ARPANET host-to-host protocol was insufficient or where inter-networking was required With DARPA support into protocol has come ARPANET and several TCP implementations were done and the its use widespread relatively is still spreading within use TCP is the scheduled to replace the ARPANET host-to-host protocol throughout the net by I January 1983 of Meanwhile the host-to-host protocol that the INWG was working on was finished PTTs and North American common 111-70 iA and documented carriers submitted rest just as the the X 25 Report No standard 4799 to Bolt Beranek and Newman Inc CCITT likely play little so the INWG consensus protocol will most operational role in the ARPANET or elsewhere I1 III-71 Report No 4799 1 4 5 Bolt Beranek and Newman Inc Network Growth -- A Summary In the following three subsections we consider tnree aspects of the growth of the network the network topology the traffic growth and the increase in the growth of the number and type of hosts on the network 1 4 5 1 Traffic Growth In early internode 1973 Roberts presented a curve of average host traffic growth for the network which showed the level of internode network traffic to be increasing factor of ten every ten months at a rate of a Internode traffic means traffic sent from a host on one node to a host on a different node i e it does not include traffic sent between hosts on the same Based on this rapid rate of growth would run out following figure of capacity in node Roberts predicted the network nine shortly after Roberts' months As shown in prediction the rate the of internode traffic growth decreased sharply to roughly a factor of two every twenty months It is interesting to speculate on the reason for this sharp decrease L G Roberts Network Rationale A 5-Year Proceedings COHPCON 1973 February 1973 pp 3-6 111-72 Reevaluation Report No 4799 Bolt Beranek and Newman Inc ARPANET HOST INTERNODE TRAFFIC I 106 W JAN 72 JAN 73 JAN 74 JAN 75 JAN 76 JAN 7Y Figure 4I- Growth in Average Host Internode Traffic 111-73 Report No It new 4799 Bolt Beranek and Newman Inc can be hypothesized that the existing hosts and the hosts that were added were used remotely more and more and network traffic increased more and least the popular time-sharing capacity this made it to get service more until the hosts began to resulted in turn traffic growth running out capacity as predicted by Roberts the hosts ran out of run Therefore capacity instead while at out of attempt in a leveling off of network of hosts pointless for new remote users to and few the of the it network seems that network still has capacity left As in already stated the traffic shown in Roberts' the figure above includes only internode traffic two reasons for excluding intranode traffic curve and There First are intranode traffic puts a burden on only one node rather than on the network as a whole calculate Thus when Roberts the effects naturally excluded intranode of for instance was attempting host traffic on network capacity intranode traffic statistics traffic Second the and there is traffic node from It significant to the same data traffic between two hosts on the same believed amount available no convenient way to separate this looped test actual is he include some amount of test traffic being looped from a host through its node and back host to of however real that traffic there is actually a between hosts on the same III-74 NE Report No node 4799 Bolt Beranek and Newman Inc For instance a week-long amounted to a daily level of intranode tr-ffic the measurement Kleinrock reports that during average of twenty percent of the level of internode traffic in some one-hour intervals the intranode traffic level was as much as eighty percent of the internode traffic level available long term statistics on inter- and intranode and forty percent of internode traffic levels curve given factor if in the figure all traffic is traffic network traffic phenomena one is First is computers in is the IMP is there is the should be scaled up by this a significant a handy are the happen even if two it of all very soon pressure computer center connected computers was not interface and a computer center to connect some host to communicate with other computers in between portion interhost to communication between the computers is computers traffic interesting and probably indicative of four installed in onto the network above Thus the to be included That intranode traffic two A scan of that intranode traffic levels have averaged between twenty shows once and to connect other the IMP so that desired possible Second when the same IMP so they may both the network communication themselves comes free and begins to initially thought to be desired L Kleinrock and W Naylor On Measured Behavior of the ARPA AFIPS Conference Proceedings Volume 43 May 1974 pp Network 767-780 111-75 f Report No Third 4799 the TIP a host has been chosen by several sites as the most flexible traffic Bolt Beranek and Newman Inc available multiplexor concentrated weak at a have that while center managers appears in fact cynics the world of also many who Once the network became some sites have arranged with some site's computer site managed it other sites that was moved to a second site and the second for the first site which used the the network via a simple terminal concentrator this some running computer centers but do so because they need the available for be to include not only managers service supplied by the computer center one to guessed that every computer center manager is trying whose inclinations are as the cynics guessed but dislike Fourth there certain site and therefore often on the same to build his empire as large as possible computer TIP-to-host but definite tendency for hosts The reason for this tendency is would and at these sites is likely to be intranode is a as yet still IMP terminal computer over A further reason tendency for hosts to be clustered is the economy of scale possible when only one facility and staff is required for the operation of several computers 1 4 5 2 The Topology first ARPANET node was installed at the University of California at Los Angeles in late 1969 and the next were installed in California and Utah 111-76 three nodes Report No 4799 Bolt Beranek and Newman Inc i UA 1 I Figure 5 S 111-77 The ARPANET in December 1969 Il By Bolt Beranek and Newmano Inc 4799 Report No I June 1970 three East Coast and two more West Coast nodes were added as well as two cross-country lines SR HARVA Figure 6 June 1970 111-78 Report No 4799 Bolt Beranek and Newman Inc IMPs continued to be delivered to the field rate of approximately one per month were thirteen IMPs installed in entirely compatible all an average so that by late 1970 there the network being at based on The IMPs the were all Honeywell 516 computer MIT Figure 7 o December 1970 III-79 Bolt Beranek and Newman Inc Report No 4799 By two-thirds of the way through 1971 IMPs had been two 516 additional installed the prototype TIP was rinning at BBN and two TIPs were operational within the network at and MITRE AMES I WCOLN S41 ANDILLINOIS Figure 8 CREI September 1971 111-80 iuN Report No The 4799 TIPs Bolt Beranek and Newman Inc were computers and had as based a on Honeywell component a 316 316-based instead IMP of 516 which was completely compatible with the 516 IMP but half as expensive early 1972 several additional IMPs and TIPs had been installed and the central part of the network between Coast clusters was beginning to fill JTAN AFG K the East and out AESILNOSEI UC8 S Figure 4 By 9 March 1972 111- 81 West Report No 4799 Bolt Beranek and Newman Inc By August 1972 a third cross-country line had been added and it was clear that in addition to the IMPs scattered throughout the center of the country in four Francisco geographic there were actually clusters areas Boston Washington of IMPs D C San and Los Angeles So McCLELLAN 89N UTNCASE AFGWCso UTAI oAMES U S MIT NOA ILLINOIS IT R TIKE UNL A RPAT Figure 10 August 1972 111-82 Report No ' 4799 Bolt Beranek and Newman Inc There follow once-yearly maps for the with v which the reader can years 1973 SIMT 4 IE I 1 Figure 11 1977 follow the continued growth of the ARPANET topology l to September 1973 III-83 oI'' Report No 799 Bolt Beranek and Newman Inc Figure 12 June 197 Bolt Beranek and Newman Inc Report No 4799 LINCOLN MIT-0q MIT-MAC CCA lox Mo AM AMES STAN EX SRI UYAN GWC ARGONNE PURR UR TYMSHARE FNWC HAWAN OCLA UCSD RAN10 US0 1 WPAFB SELVOIR SDAC SOAC MITRE APA ILLIWIS AFWL us NYU ASEROM IWRSAR N ETA GUNTER EGLIN Usc-ISI RML 0 IMP s TIP SATELLITE CIRCUIT Figure 13 July 19T5 111-85 LONDON Report No J4799 Bolt Beranek and Newman Inc OTPO TAWM PU SuS fWMEXTIIMaOESmT@wMFgVFIha vSCC Beranek and Newman ReportBolt lnc flifuAkk'$ fPS x 4 b qrt WC Ors#t 01FiSO C F4CIOIV UAiCS NUE UECSSM flNOB SOWW1 1IM #4 NAMS SATCLT Fi4r 15 V111-8 e silts Jul 1977 Report No 4799 Bolt Beranek and Newman Inc Having shown the growth of the ARPANET series is it maps geographic of topology through interesting to consider the evolution of the topology on a more quantitative basis the topologies given in and 7 NORSAR are missing the first maps were made the Figure 16 chart six entries and LONDON nodes did not earlier We exist The first at Columns because the HAWAII the times three entries in the early days of the network this nodes has stayed about note that intranode throughput actually is length reached in path 1974-75 First the number the same since 1975 the network throughput has continued of total network throughput kept Also to increase Next a significant fraction note the peak in average path and the subsequent decrease in average 1975-76 as a direct response to network delay problems which occurred from Finally HLWAII the NCC in Barbara in note that in to LONDON is July 1975 and July 1977 was 15 hops from and FNWC Monterey 1977 SRI2 near San Francisco NRL near Washington in the path equaled by other paths in the network the Boston area California of Despite length selected lines were added to the network in 1974-75 in There are several interesting facts that should be noted from the chart network six columns 8 through 11 are missing because that information was not the use the eleven maps shown above as the basis for the quantitative data shown in 6 a California California D C 111-88 both is UCSB in Santa 1975 and eleven hops from Report No 1 DEC69 4799 Bolt Beranek and Newman Inc 9 10 11 21 3 I 4 1 51 6 1 71 8 1 ----------------------------------------- ---------- -------41 2 001 1 331 21 I I I --- III I - -- --- JUN70 1 91 2 221 2 311 41 DEC70 1131 I I 2 461I 2 761I 61II --- SEP71 1181 2 441 3 321 71 --I I i -- I II 1-13 27% 2 8921 3 1211 6 013 --14 00%1 11 6331 21 0731 32 706 AUG72 1291 2 211 4 68 91 --- 1-11 79%1 682 5021 287 9531 970 455 I I MAR72 1231 2 351 5 041111 I III III SEP73 1401 2 201 5 611131 5 4011113 53%1 2 893 1301 I I I II I I 742 7461 3 635 876 JUN74 1461 2 171 6 141131 5 9811211 19%1 3 125 9551 1 513 7771 4 639 732 I III III I JUL75 1571 2 281 6 791151 6 681151 67%1 5 179 3611 1 018 5381 7 097 899 JUL76 1581 2 451 5 261111 5 131101 5851 111 2 411115 371111 'It 1 5 271111 1 1 9451 JUL77 158 6 627 9681 1 961 7261 8 589 694 7 051 9221112 706 054 9 757 976 where the columns contain the following information Column Column Column Column Column Column Column Column Column Column Column 1 Date of Map 2 Number of Nodes 3 Average Connectivity 4 Average Path Length Maximum Path Length 6 Average Path Length Minus 3AWAII NORSAR 7 Maximum Path Length Minus HAWAII NORSAR 8 Percentage of Node Unavailability 9 Internode Throughput 10 Intranode Throughput 11 Sum of Internode and Intranode Throughput Figure 16 Some Quantitative Data 111-89 LONDON LONDON Report No 4799 There has Bolt Beranek and Newman Inc been a behavior of the ARPANET good deal of actual measurement of the and the most detailed discussion of this is presented in the paper entitled On Measured Behavior ARPA Network L Conference Proceedings Kleinrock Vol and 43 pp 9 I W E 767-780 Naylor of AFIPS the 1975 Report No 1 4 5 3 Bolt Beranek and Newman Inc 4799 Hosts an SDS-940 at SRI at UCLA SIGMA-7 an IBM 360 75 at UCSB DEC PDP-1O at the University of Utah indication of the diversity various hosts of map schematic recent the host manufacturer ILLIAC-IV relatively and a types and There follows a ARPANET Figure 17 showing the Notice that the hosts range from small PDP-lls to large IBM systems with a scattering of very special as SDS This beginning gave a go d which would use the ARPANET systems operating of an were four hosts connected to the ARPANET The first a running standard variety ones provided of systems operating by hosts manufacturers to such from very special ones constructed by university researchers Figure 18 provides a breakdown of the numbers and kinds of hosts on the network NIC for inclusion in share a single twenty-three TIPs based on information compiled by the the ARPANET directory host in is It port the on an ARPANET IMP hosts Some of these Also logically each provides of a the host function although no physical host separate from the network node is required A sampling of such schematic maps covering the of the ARPANET is provided in Appendix B 111-91 entire history Bolt Beranek and Newman Inc No 4799 -Report FW MOFFET o FOPL-11 HFigure 1 T IMP EA L CUTAH P LL POORAT1O ILINOS CAM-1 ca -aa A LE OP-11 PO -1 NOSYCCOUPSITER COIPIGIRAIO CENTER1 S17 DE-100ARPCE O- C WTREas A MCAC UR C SPOPUE BY HE M ET 10 Logical Nap-- 111-92PP-1 une197 Bolt Beranek and Newman Inc Report No 4799 Mr r4A M q 4 IV 4 C 0 f-in 04040-1a V r40- - Cqr4 04 #4 rq0 10 0 r-4 iO in 0 I Cn MCa424 u Q 004000 N N 1-1 0000 00 000 D 0O in' 4 0 14 4 0 0 il 1 0 -4 -40 tnoo00MC ooT no r4Mr44 C A0 r 4r 904 %%NIII 1% NI rqviAr 4 - HC'4 4A r4 in r4 r4 yf % 0 I 40 14 000 00 0 t0 aO0 W 0 62 a 000 $4 02 W V0 0 W '4 Dt 01 230 304 0 00 M 0 000 44 '7 -444U o CO a 0 M I VA-N% 40 9 4 U0 4r0 tin 000'%' 0 0 e D 0 41 s 0 03 44 4 ti S qMr 40r 4i C to%00 m4 1 00000@ 44 4r r4e A4 P40 0 i E14Ca0 H 111-93 0 A E4 P 40 0 00 d 1VV 40 n 04 4 J0 4 a an 1 111 1 1 1 filigrgigml ag lagaijii dizilzi g an at 1 a I til-I 6 1' I ll 'l'llilvtl I a 3 a 1 Report No H799 Irerll I'll t I ll L31 tin-ll Report No 1 5 4799 Bolt Beranek ana Newman Inc The Impact of the ARPANET The ARPANET testbed for a has new served well computer its original function as a communications technology More recently the ARPAKET has also given good operational service to a number of users who have come to depend on it communications service plan for However for their computer a part of the original the ARPANET was technology transfer was thought that the transfer technology would occur in of For instance interactive three forms computer technical literature 2 Through open and 3 scientific transfer and Through the military command and control centers for which the national Military Center network the common carriers or other commercial organizations concerned with data dissemination it 1 Dissemination cf techniques and experimental results through the and program Command System in the Pentagon serves as the focal point greatest siccess has perhaps been in this area Support The ARPANET's of technology transfer Being an unclassified effort implemented for the most part by individuals with an academic or research leaning naturally sets of been papers development were numerous papers written on the ARPANET written a set relatively of early in the have Two key XRPANET five presented in a session at the AFIPS 1970 Spring Joint Computer Conference and I there 111-95 another set of Report No five 4799 Bolt Beranek and Newman Inc presented in a Computer Conference and the two sets session at the Both these sessions were organized by and already listed a these and distributed number bibliographies provided elsewheie in Becker and Hayes of widely other We papers this history In Inc of Los Angeles Bibliography Becker and Hayes lists Inc 1975 is AD-A026900 1 8 5p February 1976 available from NTIS California index The number the bibliography was There was also a large number of working papers distributed among the various groups and also covered in the Becker Many computer communications of and Hayes bibliography National Bureau of Standards has also constructed a on ARPANET accession individuals working on the ARPANET development are IPTO This bibliography under A complete set of the papers in also collected on microfiche informal the and Index to P'ublications about ARPANET 561 relevant documents and includes a stlbject document have in to prepare a bibliography of publications related to the Selected IPTO of five papers were specially bound together under DARPA-provided covers commissioned AFIPS 1972 Spring Joint which includes these The bibliography hundreds of suggested the PRPANET-related publications In mid possibility Spring Joint instead 1971 Robert Kahn then at BBN of h public demonstration of the ARPANET at the 1972 Computer Conference Lawrence for this demonstration to be held in 111-96 Roberts arranged October 1972 at the Report No first 4799 Bolt Beranek and Newman Inc Internaticnal This Conference on Computer Communication demonstration development thoroughly It forced debug protocols It marked all key participants gave international by several the key in the visibility communications members Fifty to community resources in ARPANET project to the packet Robert and Kahn personnel and of the ARPANET community planned and managed the full year effort which demonstration point until then had been viewed largely with undertook the task of marshalling with turning and test their network support and application switching concept which scepticism a ICCC kilobit culminated phone lines in a were successful leased from existing network sites to the conference site at the Hilton Hotel a demonstration room in the hotel for the duration of the conference and an ARPANET TIP was set up in members of the ARPANET community were involved all manner of Washington Dozens of Manufacturers of computer terminals were invited to connect their terminals to the demonstration TIP Throughout the conference hours each day of the conference there were individuals available to demonstrate the use of programs on ARPANET hosts from the manufacturer-provided booklet was written confaeence many copies made A relatively thick available at the by means of which visitors to the demonstration area could follow do it network and terminals hosts A yourself directions to use programs on many twenty or thirty minute motion picture about III-97 Report No 4799 the ARPANET produced Bolt Beranek and Newman Inc and the promise of computer and shown at the conference TIP had not been available for a very limited communications Coming at a time when the long time when only number of terminals had been tried with the ARPANET when many hosts had completed the fnitial implementation necessary host software but few had had it the was ICCC demonstration provided of a and the running for very long an important stimulus for the ARPANET community to pull together and get the network operational was a spectacular shape The demonstration itself success with everything working amazingly well remarked that the ARPANET technology really is this impression back home with them in many visitors real and carried The assurance with which Roberts promised the demonstration and the routine way he spoke of it while it impression taken home by efforts and feelings in which was happening no doubt enhanced the the of true visitors panic of and belied the crash the members of the ARPANET community who were called upon to execute the demonstration There have technology Of been numerous course many other of these relatively low key or relatively short the demonstrations example IPTO community demonstrations several the have been informal or but in several have been truly major productions with help from of members of the instances Once for ARPANET took a demonstration team to Europe where a series of demonstrations was given for members of the NATO community over a two-week period 111-98 Report No 4799 Bolt Beranek and Newman Inc There has been good technology Defense to other success parts Communications of in transferring gaining Agency experience PWIN and EDCN networks procured AUTODIN I two small function the first The networks for the purpose with the ARPANET technology Called the of these was used in effort and the second was used in to ARPANET the Department of Defense essentially identical to the ARPANET in of the connection with the WWMCCS the successor Two other networks essentiaily identical to the ARPANET but smaller have also been procured by other parts of Where direct copies of the ARPANET were not appropriate the DoD ARPANET technology has nonetheless affected the characteristics of new DoD networks being common user network For built AUTODIN II the instance new being constructed by DCA DoD is explicitly a second generation ARPANET Outside the U S use the companies ARPANET military technology the commercial world has begun to or variations have filed with the F C C on it Several or already been licensed to offer communications services based more or less directly on packet-switching technology developed in are Telenet Communications Corporation for the financing and Lawrence Roberts was Now with GTE Subscriber Network Products 111-99 ARPANET the Among these BBN arranged President Tymnet which Report No 4799 Graphnet IT T Bolt Beranek and Newman Inc and AT T ARPANET expertise used the about of the several ARPANET eighty to in ARPANET Tymshare as a substantially and AT T all continental a wide Telenet now serves U S geographic although those have has and has made several foreign networks and serve in Tymnet developed announced area related a new version of to Telenet parallel of making the time-sharing services of different closer carriers Tymnet initially developed to technology however IT T to means available techniques the connect several foreign cities to common technology most directly cities arrangements Because of its access to substantial and a communications being built for ARPANET their used intention uses Graphnet to provide public packet-switching services A number procuring of private techniques U S companies corporate networks developed for ARPANET announced that Citibank of New contract to BBN have also procured utilizing For instance York City has many it or are of the was recently constructed by a private network very similar to the ARPANET An increasing number of commercial RFPs call for packet switching or for functions switching A which number of can be provided using packet companies have taken advantage of the fact that the ARPANET technology obtain only is in the the listings of the ARPANET software 1II-100 public domain to Report No 4799 Several Bolt Beranek and Newman Inc PTTs the foreign national Postal Telephone Telegraph authorities have made a commitment to the of packet-switching research networks developed or are networks there and development have been several foreign and several international networks under consideration are being There follows a list of some of these networks CIGALE -- an operational network developed by a French government research agency RCP -- built by the French PTT and operational as a teatbed TRANSPAC -- under construction by the French scheduled to become operational for public use in EPSS -- an and 1978 experimental packet-switching service built by the UK PTT which became operational in CTNE -- PTT 1976 a packet-switching service operated by the Spanish PTT Datapac -- a packet-switching network Trans-Canada Telephone System which is in the built early by the phases of operation ------------------- More detail on this list may be found in Planned New Public Data Networks P T Kirstein Computer Networks Volume 1 Number 2 September 1976 pp 79-94 Kirstein is himself a member of the ARPANET community and was instrumental in arranging the installation of the London node of the ARPANET III-101 S Report No 4799 JIPNET EIN -the -- Bolt Beranek and Newman Inc practically a copy of the'ARPANET built in the European Information Network U K Switzerland France built and Italy Japan jointly by and strongly the European influenced by CIGALE EURONET a network Eco'romic Cmmuri No doubt therf' under discussion by initially to use the EIN technology are other networks besides those mentioned above in the planning phase or under construction networks coming into become important issues representatives in being to From the time of the 19 72 ICCC cf various countries and institutions ipterested consider possible standards Network Working Group INWG Network many technology exchange and standards computer networks met informally to discuss and With so Working selected to be Group INWG's their eiperience In 1972 the International was formed Modeled on the yARPANET Vinton Cerf of the ARPANET community was first chairman and DARPA offered the services of the the ARPANET NIC to coordinate and distribute INWG working notes Later INWG became International Federation for Information back its its NIC support beginning those used in Processing with the DARPA cut and DARPA's role in INWG decreased INWG was a forum at which the associated ARPANET were other considered nonetheless ARPANET for a long time remained the one against which new ideas were compared 111-102 techniques big existing From than the network The Bolt Beranek and Newman Inc 4799 Report No international Date'ac of urging CCITT the international PTT UK the and Telenet standardization efforL the With effective has been especially Transpac packet-switching communications standards organization at which all of the world's communications authorities are represented speed remarkable with adopted standards for connecting hosts to packet-switching X 25' and X 75 these standards clearly address importance in the for purposes of issues which were first international communication development ARPANET as Known networks to each other networks and packet-switching of has seen to be perhaps where the ARPANET was seen to be deficient While today the operational primary transmission stream of packet-switching ARPANET research is purpose of of traffic the it is ARPANET in still the the main example For development network is underway to develop - more efficient and loop-free rcuting algorithms - multi-destination broadcast and group addressing capabilities improved - networ k flow control and congestion control mechanisms host attachment to more than one node - III-103 o lel l I n io i lu i11l I lin li lm t il II i i i Il II A Report No 4799 - Bolt Beranek and Newman Inc logical host naming i e to permit a single host to have more than one logical name - host interface enhancements routing 11-lO0 to permit internetwork Report No 4799 1 6 Bolt Beranek and Newman Inc Maturity and Handover to DCA A memorandum of agreement was worked out between DARPA and DCA for management of the ARPANET to be transferred to DCA as July 1 1975 with until December 31 with ARPANET a of six-month phase-over period from July 1 during which DARPA would continue to help DCA management while DCA acclimated itself to the job The memorandum also called for a detailed transition plan to be written which was completed by June 1975 Along DCA with the transfer of ARPANET management from DARPA to the technical functions that had been being performed at RML were also transferred to DCA and the procurement transferred from RHL to DECCO which is functions were a field activity of DCA There are several aspects of the memorandum of agreement and the transition plan network which worthy of invented of operation sponsors NBS who originally owned ARPANET equipment it was turned over to was by DCA DCA before Ownership of the was to finance and maintenance of the network through use of the DCA managed Communications Industrial Fund costs The to be used solely The concept of ARPANET equipment was to remain with the sponsors its here with sponsors being those users or collections of users DARPA management the mention was to be an operational DoD facility for government business e g are a pro-rated which would recover allocation to sponsors based on the 111-105 Bolt Beranek and Newman Inc Report No 4 799 equipment used by the sponsors with BBN and DCA was perform SRI to NIC functions network miitr maintenance and topological consulting and after certainly and contract ARPANET with NAC initially operations initially and if needed it was clearly implied that was DCA could retain other contractors eventually the to to the perform first these year functions DCA was to and thereafter if operate the network for a period of three years be provided on a necessary until equivalent service could S111i10 o - Report No 4799 2 Bolt Beranek and Newman Inc OBSERVATIONS For many contractors centers of the and in people in at the participating universities the and major research the development of the ARPANET has been an exciting time which will rank as a high point in In government their professional 1969 the ARPANET project represented a high risk high impact practical research useful effort form has The not existence only of provided technology to meet many short term needs but careers potentially the net in communications it represents a formidable communications technology and experience base on which the Defense sectors will strong and Department as well as the entire public and private depend diverse for advanced experience needs communications base generated The by the ARPANET project has placed this country ahead of all others digital communications science and technology 111-107 boo in advanced Report No 2 1 Bolt Beranek and Newman Inc 4799 Social Issues Somewhat change in become expectedly the network has facilitated a social the United States computer research community more convenient for geographically send The files of text between geographically ability ea3ily and ability to collaboratively use powerful editing and document production facilities ollaborative change in has chauged research improvements in a to remote groups the ability to'communicate via messages quickly and the has separated groups to perform collaborative research and development easily It significantly with remote groups human communication in the rate of the feel Just as other major the past have resulted progress of in this social effect of the ARPANET may finally be the largest single impact of the ARPANET dcvelopment A non-trivial country is would question of considerable importance to the why the ARPANET project has been certainly be nice if so successful It the same formula could be applied to other pressing national needs While timing accident must is to identify several not be discounted it possible and luck possible contributing factors o Despite the fact project and that further the ARPANET despite within the Defense Deoartment 111-108 i was a government the fact that it it was possible was run at least Report No 4799 Bolt Beranek and Newman Inc initially some of other to free the the constraints activities unclassified was for a as to free make cost benefit government unauthorized possible early to of project provided the and to entirely government without probably Department the a ARPANET about it without Finally did not have to interconnect directly with other existing communication systems it was possible to explore line protocols and interface standards dL in the best ways ARPANET program requirements was complex administrative procedures for access control the valid concern facilities ARPANET being ill-advised despite government build was network Third Defense use from good this allowed people to decisions and the time experiment with the use of forced development the provision of network service long a and which often seriously hamper First Second very contractors research that was could incredibly be devised free of nova Thus the artificial and was able to concentrate intensively on the primary required research later in after success had been assured was the project relatively and development it easy to reopen consideration of some of these issues and at the current time Defense Much Communications Agency 111-109 is for example the charging user groups Report No 4799 Bolt Beranek and Newman Inc for network access using private and there have been experiments line interfaces in the transmission of classified data over the ARPANET and research was on how to implement network login procedures o A very convenient fact was the common DARPA done etc support etc of both the network authority and the initial early group of network users encourage a engineering It was possible for DARPA to strongly cooperative at the time attitude in the and cooperative project when such cooperation was most critically necessary In sum the accomplished management project with was strong adequate an illustration technically resources and of what can sophisticated be central a clearheaded undeviating concentration on the central research and development issues The largest single surprise of the ARPANET program has the incredible popularity and success of network mail little doubt that the techniques of connection country and with the ARPANET drastically intercommunication in program change the network public the and are mail going techniques private There is developed used the telephone for sectors primary prior existing communications techniques the U S and in to sweep the hindsight one can easily see the reasons for this success service been By The postal have certain serious deficiencies III-110 Report No 4799 Bolt Beranek and Newman Inc The postal service has become more expensive and its in measured in telephone days for delivery of letters In the case of the our increasingly mobile society makes it difficult to person busy people are hard to reach in any reach the case and reaching 10 distributed people within a of desired time works if performance is essentially impossible a careful individual checks frequently but it system back is is short period Leaving telephone messages established and a travelling with his answering service or secretary often inconvenient and is usually limited to the prime shift working day of the secretarial world To find a busy person able to accept a phone call at the time you make it is little for truly unusual and some officials have desks covered with pink slips reporting on telephone messages return calls with requests Into this milieu was dropped a technique of network mail where at any time of the day or night one can send a message to any number of other people and expect that message semi-instantaneously the recipients part of all be available in the computer mailbox of all Then one only has to assume the habit on the individuals using the system to occasionally check their mailboxes when they are free and not at a meeting performance to of the communications represents of messages forwarding messages sophisticated With tools for answering messages and III-111 categorizing the an immeasurable improvement over the postal service or the telephone addition and the filing messages the Report No 4799 Bolt Beranek and Newman Inc system takes on yet another step function of performance over the alternate possibilities In the space of just a couple of years this computer center curiosity became a smashing success on the ARPANET there was a sufficiently large community of who wished to communicate and they all were relatively mobile and the system overnight became a way of life of this kind of success are enormous would have eventually come anyway the individuals The implications Perhaps such advances but there is no question that ARPANET program provided a truly convincing demonstration of the power of this approach because it does depend The change will upon the not availability accessible to wider and wider populations be of overnight terminals but commercial systems are already available and substantial DoD experiments are already under way 111-112 Report No 2 2 Bolt Beranek and Newman Inc 4799 Technical Lessons Leaving the broad social plane the ARPANET program provided several technical lessons which are worthy of general comment 111-113 Report No 2 2 1 4799 Bolt Beranek and Newman Inc Terminal Handling Rather early in terminal access inadequate terminal the ARPANET program to the approach access locations many classes of users contractor designed to that required direct response to this pressure was the design of Bolt address character-oriented clear order to use major hosts at other the Terminal Interface Message Processor prime became net via the main host computers was an to the net in The first it Beranek a and by Newman limited asynchronous TIP the Inc problem terminals ARPANET The TIP was to only and handle to be an integral part of the network authority and not available for user programming or special user features This limited goal and hard-nosed attitude permitted the rather rapid completion of the TIP design the fielding of many TIPs of widespread terminal access to the effort was extremely successful in Unfortunately goal and absolute considerable community terminal Justified terminals but perhaps restriction unhappiness in and the rapid availability network reaching its not on user portions of this techniques approach had leapfrogged TIP the limited programming the the limited goal surprisingly and created considerable pressure for access Thus created potential other user better Some of the complaining was fully to servicing interactive character a whole segment cf the industry that I11-114 Report No 4799 Bolt Beranek and Newman Inc was concerned with batch processing and the disciplines line from such batch use of processing synchronous units complaints were less rational and more self serving Other where some groups really wanted a computer of their own under the guise of obtaining terminal access to the ARPANET Still other criticism was based on an honest difference of opinion as to ease of designing and the relative deploying a terminal access device that would permit user programming and would handle a broader class of terminals Finally sophisticated users some who criticism were was used to provided the by terminal services provided by the most advanced large hosts highly support and did not like the limited services provided by the tiny mini-host resident in the TIP In response to this pressure DARPA for a time supported the development of a device called the ARPA Network Terminal System ANTS to This was a system based on a PDP-11 which provide user programming ability of services to the terminal user were somewhat ambitious and the field the difficulty in delays in debugging and a more Unfortunately although a few ANTS configuration fielded intended the handling of synchronous terminals as well as asynchronous terminals set was management versions of of were powerful the goals put the program the system into the and implementation eventually led to a cessation of DARPA support for the effort In response to a somewhat different 111-115 set Report No 4799 of Bolt Beranek and Newman Inc ELF attempt was made with a system called another requirements Here DARPA support was provided to improve deploy and a PDP-11 try Again the efforts to goals were far more limited orderly for a particular the hope was to permit user programming and the handling of a wider class of terminals deployment standardize based terminal support software system which had been independently developed project to development the From viewpoint use ELF were much more successful and more and maintenance attention However was paid specialized host functions than as a many TIPs in the network First it is There are extremely machine Second configuration is central program an an bit like the to software design and release close control of evolving computer-based device is to be deployed in considerable numbers around the still a very great attention must be paid if this and leads to an unlimited expansion of the management changes to difficult to build a system central maintenance and software support and program morals several which can handle all possible terminals it problem primary terminal either through the main network hosts or via the access was still everything users terminal the time the ARPANET was transferred to DCA history serving means of widely distributed network access for large numbers of At to ELF has probably been more useful as a base for individualized mini-hosts local of network There is open technical question whether such a device could be 111-116 Report No 4799 Bolt Beranek and Newman Inc sensibly and cost-effectively fielded in a way which kwgld permit both some level of user programming to tailor the device and this point in approach the functions of a standard set of controlled basic services at the development of the precludes such user technology programming if a conservative a device is to be widely deployed Another related aspect of the terminal handling problem to do with the management extent of the network authority was discovered that when terminal in has an unsophisticated user typed It at a a remote location and something went wrong that is the proper response was not received the user typically blamed the network despite the fact that any number of possible things the terminal itself at the local modem the other end of the local line e g any TIP the network itself similar Thus it is adequate the local line the eventual host over it is to be in numerous authority host crisis incidents During the the computers network authority Perhaps III-117 ARPANET were precipitated by the failure of equipment which was not in any way under of have a position to respond to such unstructured outcries of rage from the end terminal user program or the entire collection of equipment from the terminal right through to the if computer could have been at fault extremely important that the network control modem the terminal handling device point on the return path administrative the the control the best single example was Report No 4799 Bolt Ber3nek and Newman Inc difficulties in the local on-base telephone system at Ames it took a massive effort where to eventually uncover the offending equipment and where the network authority was forced in its own defense to participate and lead in the massive debugging activity even though the outside the lesson network from networks offending device eventually found was clearly authority's control The generalizable this kind of history is that devices that attach to must be responsibility extremely Either the clearly device camp in- somebgdy' 1 of must clearly belong to the network authority and must have built-in techniques for debugging and failure location host organization understands its l' cation or some other responsibilities devices authorities or the device must clearly cannot sit well for in belong identified maintenance cracks the group who and between Although this idea is really quite to trouble different simple it is frequently overlooked A final point on the general terminal handling problem has to do with the location of the intelligence terminal related matters In general can either be in the terminal itself in network port e g or way of course the TIP for is with the data processing power the terminal handling in the eventual final host computer distributed among these three locations in some As the cost of data processing is dropping intelligence dealing appearing in the terminal itself 111-118 more and more and this entire R9 Bolt Beranek and Newman Inc the reviewed to ensure that periodically be must matter of technological change performance can take advantage 1- III-119 system Report No 2 2 2 4799 Bolt Beranek and Newman Inc Reliability and Fault Isolation At the outset of the ARPANET design considerable expended toward recovery For built-in techniques example IMP-to-IMP patched for fault isolation recovery mechanisms neighbor enough circuits it was therefore techniques for of failures to add techniques fault isolation The lesson probably is interesting cross power fail somewhat of a techniques and there was a nearly improve rapid recovery that it too much attention to fault isolation Many be efforts down this path were not nearly Over the life of the ARPANET program effort could and and mechanisms for reloading one IMP from a that the initial continuous isolation and IMPs themselves had As the network grew surprise of- fault effort was were and is and containment difficult to have recovery slowly existing mechanisms added to improve the network's ability to cope with trouble o For critical pieces computation was the operated checksummed of code code itself Similarly before they in these key data were particular could cause network-wide 111-120 the routing structures accessed was were This kind of discovered classes failures local difficulties as was checksummed before it protection was added when it trouble such that the of computations rather than simply Report No 4799 o It Bolt Beranek and Newman Inc was discovered terminals that dial-in modems for remote could break or hang and there %as no simple way to discover this designed had wherein a happened A centrally technique located controlled by the network authority used was autodialer an out WATS line to periodically dial and test 311 dial-in ports all over the network at least in the continental United States o It was discovered that when a difficulty occurred in IMP which caused an automatic program an reload the necessary debugging information was lost and the trouble would likely recur automatically attempted dump A technique offending was added to code before a reload was which then permitted comparison with a master copy and improved capability for debugging o The ARPANET program simultaneous need was for forced a to cope twenty-four with hour continuously operating reliable system and at time relatively constant levels of growth and change the the a day same modification After some early false starts when large changes introduced periods of substandard performance a technique was evolved whereby greater attention was paid to dividing a large change III-121 into small incremental Report No 4799 Bolt Beranek and Newman Inc changes Then that when were compatible with the previous system trouble concentrated occurred debugging could be on the small incremental change or retreat taken to the previous release o As network users began to depend upon a service hosts for a wide including message services hat such hosts In some cases it few variety became maintain availability a host might be operating particular of services more important to the network adequately perceived by its own local operators and yet in not be properly technique was added servicirnZ evolved software the network connection whereby the network critical hosts and was then corrective by the local host authorities if 111-122 watch in a these position A authority to when necessary ki' some way tools action as particular to urge and Report No 2 2 3 f 4799 Bolt Beranek and Newman Inc Maintenance Management Inthe early years of the ARPANET program the IMPs and TIPs of the network were maintained by subcontract to the manufacturer of the basic mini cost-effective computer approach Honeywell because facilities in many cities However computer simply maintenance reliability was requirements of This Honeywell represented had maintenance this rather standard form of inadequate the for ARPANET the After high a time maintenance of' the network nodes was undertaken directly by Beranek and Newman Inc the network prime nodal reliability In particular hardware close personnel became highly expert in concert with a small itself behavior it an offending node and number to could difficulties with the local maintenance engineer people ARPANET in of field observe talk the through Further the the field became much more dedicated and responsive to difficulties as compared to personnel who had to take responsibility of team the IMP and TIP machines The central staff could use the network of the and s3ftware experts at the central Network Control Center acted in who and techniques of central maintenance management were developed wherein a very strong of Bolt contractor special new techniques were developed which greatly improved average a equipment in their geographical time-shared Honeywell for many different kinds territory This approach to maintenance probably has important benefits for other distributed 111-123 Report No systems 4799 Bolt Beranek and Newman Inc especially interconnected in as those networks and systems thereby maintenance III-124 4 ---- z - - ' ' will increasingly be accessible to central Report No A A 1 4799 Bolt Beranek and Newman Inc BBN ARPANET Bibliography Technical Reports AD682905 BBN Report No 1763 Message Processors January 1969 ADAO19160 BBN Report No Sinterconnection of lnitial for the Design for Interface ARPA Computer Network 1822 Specifications for the a Host and an IMP revision of January 1976 AD730725 BBN Report No 2161 A Study of the ARPA Network Design and Performance Kahn and Crowther August 1971 ADA014398 BBN Report No 2183 Terminal Interface Message Processor User's Guide updated version of August 1975 j ADA002481 BBN Report No 2184 of November 1974 AD776995 BBN Report No 2277 Specifications for the Interconnection of Terminals and the Terminal IMP Rettberg revision of June 1973 TIP Hardware Manual revision BBN Report No 2491 Throughput in the ARPA Network S-Analysis and Measurement McQuillan January 1973 text also contained in QTR No 16 January 1973 ADA025358 BBN Technical Information Report TIR Interface Message Processor Program of June 1976 Version 3231 No 89 The updated version ADA033351 BBN TIR No 90 The Network Control Center Program updated versiQn of November 1976 Version 131 ADA031617 BBN TIR No 91 The Terminal Interface Message Processor Program updated version of August 19T6 Version 400 AD781467 BBN Report No 2831 Adaptive Routing Algorithms for Distributed Computer Networks McQuillan May 1974 A-1 Srl i llll l I 11 lll iAr Report No 4799 Bolt Beranek and Newman Inc BBN TIR No 93 The Remote Job Entry Mini-Host August 1974 text also contained in QTR No 6 July 1974 BBN Report No 2891 A Proposed Experiment in Packet Broadcast Satellite Communications Rettberg and Walden September 1974 BBN Report No 2918 Network Design Issues November 1974 text also contained in QTR No 7 October 1974 ADA022040 BBN Report No May 1975 2999 Pluribus Document 1 BBN Report No 2930 Pluribus Handbook January 1975 BBN Report Configurator Document No 3000 Pluribus in production BBN Report No 3001 Pluribus Software December 1975 BBN Report No 2931 Software April 1975 2 Document Pluribus Document 5 ADA018341 No 3004 Pluribus September 1976 System Document BBN Report No 3002 Pluribus Document 6 Specifications February 1976 BBN Report Maintenance Overview 4 3 Basic Advanced Functional Document 7 BBN Report No 3056 The Atlantic Satellite Packet Broadcast and Gateway Experiments R Binder R Rettberg and D Walden April 1975 BBN Report No 3126 A Multiprocessor Design W Barker September 1975 A-2 Report No A 2 Bolt Beranek and Newman Inc 4799 Published Papers W Teitleman and RE Kahn A Network Simulation and Display Program Proceedings of the Third Annual Princeton Conference on Information Sciences and Systems March 1969 F E Heart R E Kahn S M Ornstein W R Crowther and D C Walden The Interface Message Processor for the ARPA Computer Network AFIPS Conference Proceedings 36 June 1970 pp 551-567 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 300-316 also in Computer Communications P E Green and R W Lucky eds IEEE Press 1975 pp 375-391 also in Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 60-76 F E Heart and S M Ornstein Software and Logic Design Interaction in Computer Networks International Computer State of the Art Report No 6 Computer NetworkS Infotech Information Ltd Maidenhead Berkshire England pp B23-462 R E Kahn Terminal Access to the ARPA Computer Network in Courant Computer Science Symposium 3 -Computer Networks R Rustin ed Prentice-Hall Englewood Cliffs N J 1972 pp 147-166 R E Kahn and W R Crowther Flow Control in a Resource-Sharing Computer Network Proceedings of the Second ACM IEEE Symposium on Problems in the Optimization of Data Communications Systems Palo Alto California October 1971 pp 108-116 also in IEEE Transactions on Communications Vol COM-20 No 3 Part II June 1972 pp 539-546 also in Advances in Computer Communications Chu ed Artech House Inc 1974 pp 230-237 also in W W Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 117-125 D C Walden A System for Interprocess Communication in a Resource-Sharing Computer Network Communications of the- ACM Vol 15 No 4 April 1972 pp 221-230 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 340-349 Ordered according to date of writing A-3 Report No Bolt Beranek and Newman Inc 4799 R H Thomas and D A Henderson McROSS -A Multi-Computer Programming System AFIPS Conference Proceedings Vol 40 June 1972 pp 281-293 also in Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 246-258 S o Ornstein F E Heart W R Crowther S B Russell H K Rising and A Michel The Terminal IMP for the ARPA Computer Network AFIPS Conference Proceedings 40 June 1972 pp 243-254 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 317-328 also in Computer Communications P E Green and R W Lucky eds IEEE Press 1975 pp 354-365 H Frank R E Kahn and L Kleinrock Computer Communications Network Design -Experience with Theory and Practice AFIPS Conference Proceedings Vol 40 June 1972 pp 255-270 also in Networks Vol 2 No 2 1972 pp 135-166 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 254-269 A A McKenzie B P Cosell J M McQuillan and H J Thrope The Network Control Center for the ARPA Network Proceedings of the First International Conference on Computer Communication Washington D C October 1972 pp 185-191 also in Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 319-325 R E Kahn Resource-Sharing Computer Communications Networks Proceedings of the IEEE Vol 60 No 11 November 1972 pp 1397-1407 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 208-218 also in Computer Communications P E Green and R W Lucky eds IEEE Press 1975 pp 537-547 J M McQuillan W R Crowther B P Cosell D C Walden and F E Heart Improvements in the Design and Performance of the ARPA Network AFIPS Conference Proceedings 41 December 1972 pp 741-754 Network Analysis Corporation University of California at Los Angeles A-4 Report No W R F E 4799 Crowther Heart Bolt Beranek and Newman Inc R D A Reservation-ALOHA Rettberg System D C for Walden S M Broadcast Ornstein and Communication Proceedings of the Sixth Hawaii International Conference on System Sciences January 1973 pp 371-374 D C Walden Host-to-Host Protocols International ComputerState of the Art Report No 24 Network Systems and Software Infotech Maidenhead England pp 287-316 N W Mimno B P Cosell D C Walden S C Butterfield and J B Levin Terminal Access to the ARPA Network -Experience and Improvements Proceedings of the Seventh Annual IEEE Computer Society International Conference San Francisco California February 1973 pp 39-43 also in Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 287-291 E W Wolf An Advanced Computer Communication Network AIAA Computer Network Systems Conference 73-414 F E Heart S M Ornstein W R April 1973 Crowther AIAA and W B Paper Barker No A New Minicomputer Multiprocessor for the ARPA Network AFIPS Conference Proceedirgs 42 June 1973 pp 529-537 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 329-337 also in Computer Communication Networks R L Grimsdale and F F Kuo eds Proceedings of the NATO Advanced Study Institute of September 1973 Sussex England published by Noordhoff International Publishing Leyden The Netherlands 1975 pp 159-180 also in Computer Communications P E Green and R W Lucky eds IEEE Press 1975 pp 366-374 R H Thomas A Resource Sharing Executive for the ARPANET AFIPS Conference Proceedings 42 June 1973 pp 155-163 also in Advances in Computer Communications W W Chu ed Artech House Inc 1974 pp 359-367 F E Heart The ARPA Network in Computer Communication Networks R L Grimadale and F F Kuo eds Proceedings of the NATO Advanced Study Institute of September 1973 Sussex England published by Noordhoff International Netherlands 1975 pp 19-33 Publishing Leyden The W R Crowther J M McQuillan and D C Walden Reliability Issues in the ARPA Network Proceedings of the ACM IEEE Third Data Communications Symposium November 1973 pp 159-160 also in Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 142-143 A-5 Report No 4799 Bolt Beranez and Newman Inc J M McQuillan Design Considerations for Routing Algorithms in Computer Networks Proceedings of the Seventh Annual Hawaii International Conference on System Sciences Honolulu Hawaii January 1974 pp 22-24 also in Computer Networking R P Blanc and I W Cotton eds IEEE Press 1976 pp 108-110 S M Ornstein W B Barker R D Bressler W R Crowther F E Heart M F Kraley A Michel and M J Thrope The BBN Multiprocessor Proceedings of the Seventh Annual Hawaii International Conference on System Sciences Honolulu Hawaii January 1974 Computer Nets Supplement pp 92-95 A A McKenzie Some Computer Nptwork Interconnection Issues AFIPS Conference Proceedings 43 May 1974 pp 857-859 E A Akkoyunlu' A J Bernstein and R E Schantz Interprocess Communication Facilities for Network Operating Systems IEEE Compute Vol 7 No 6 June 1974 pp 46-55 F E Heart Implications of the Computer-Communication Partnership in Preprints of Papers MEDINFO 74 The First World Conference on Medical Informatics Stockholm Sweden August 1974 pp 21-27 proceedings to be published by North-Holland The Netherlands F E Heart Networks and the Life Sciences the ARPA Network and Telenet Federation Proceedings Federation of American Societies for Experimental Biology FASEB Vol 33 No 12 December 1974 pp 2399-2402 also in Computers in Life Science Research W Siler and D A B Lindberg eds FASEB and Plenum Press 1975 pp 209-215 J M McQuillan The Evolution of Message Processing Techniques in the ARPA Network International Computer State of the Art Report No Maidenhead 24 Network England pp Systems and Software Infotech 541-578 R H Thomas JSYS Traps - A TENEX Mechanism for Encapsulation of User Processes AFIPS Conference Proceedings 44 May 1975 pp 351-360 State University of New York at Stony Brook A-6 Report No Bolt Beranek and Newman Inc 4799 A R D Bressler M F Kraley S M Ornstein W R Crowther Michel and F E Heart Pluribus -- A Reliable Multiprocessor AFIPS Conference Proceedings 44 May 1975 pp 551-559 F E Heart S M Ornstein W R Crowther W B Barker M F The Pluribus A Michel Bressler and Kraley R D Multiprocessor System in Multiprocessor Systems Infotech State of the Art Report Infotech International Ltd Maidenhead Berkshire England 1976 pp 307-330 W R Crowther F E Heart A A McKenzie J M MeQuillan and D C Walden Issues in Packet-Switching Network Design AFIPS Conference Proceedings 44 May 1975 pp 161-175 also in Cotton eds IEEE Computer Networking R P Blanc and I W Press 1976 pp 182-196 R S Tomlinson Selecting Sequence Numbers ACM SIGCOMM-SIGOPS Interface Workshop Communications March 1975 pp 11-23 Proceedings of the on Interprocess J M McQuillan and D C Walden Some Considerations for a High Performance Message-Based Interprocess Communication System Proceedings of the ACM SIGCOMM-SIGOPS Interface Workshop on Interprocess Communications March 1975 pp 77-86 High The Evolution of a Walden S M Ornstein and D C Performance Modular Packet-Switch Conference Record of the 1975 International Conference on Communications June 1975 Vol I pp 6-17 to 6-21 R D Bressler M F Kraley and A Michel Pluribus a Multiprocessor for Communications Networks Fourteenth Annual an the mid-70's Computing in ACM NBS Technical Symposium -- Assessment June 1975 pp 13-19 D C Walden Experiences in Building Operating and Using the USA-Japan Computer Proceedings of the 2nd ARPA Network Conference Tokyo Japan August 1975 pp 453-458 R D Rettberg and D C Walden Gateway Design for Computer Networks Communications in Interconnection Network on Conference Computing European the of Proceedings Communications Networks London England September 1975 published by Online Conferences Limited Uxbridge England 'p 113-128 A-7 Report No Bolt Beranek and Newman Inc 4799 M F Kraley The Pluribus Multiprocessor Digest of the 1975 International Symposium on Fault-Tolerant Computing Paris France June 18-20 1975 p 251 abstract only B P Cosell J M McQuillan and D C Walden Techniques for Detecting and Preventing Multiprogramming Bugs in Minicomputer Software J R Bell and C G Bell eds North-Holland Publishing Co 1976 pp 301-308 A A the McKenzie The ARPA Network Control Center Fourth ACM Data Communications Symposium Canada October 1975 Proceedings of Quebec City pp 5-1 to 5-6 B P Cosell P R Johnson J H Malman R E Schantz J Sussman R H Thomas and D C Walden An Operational System for Computer Resource Sharing Proceedings of the Fifth Symposium on Operating System Principles Austin Texas November 1975 pp 75-81 R D Bressler and R H Thomas Design Issues in Distributed Computing in Distributed Systems Infotech State of the Art Report Infotech International Ltd Maidenhead Berkshire England 1976 pp 215-222 V Cerft A McKenzie R Scantlebury and H Zimmermann Proposal for an International End to End Protocol ACM Computer Communication Review Vol 6 No 1 January 1976 pp 63-89 W F Mann S M Ornstein and M F Kraley A Network-Oriented Multiprocessor Front-End Handling Many Hosts and Hundreds of Terminals AFIPS Conference Proceedings 45 June 1976 pp 533-540 P J Santos Distributed Technical 143-148 J M Software Computer Instrumentation Networks Symposium--Directions McQuillan A Status Report for Maintainability of Fifteenth and on Annual Challenges the ACM NBS June 1976 ARPANET pp Australian Computer Society International Federation for Information Processing Joint International Symposium on Data Communications Technology and Practice August 1976 pp 13 1-13 7 Defense Advanced Research Projects Agency Institut de Recherche d'Informatique et d'Automatique National Physical Laboratories A-8 U K France Report No 4799 Bolt Beranek and Newman Inc J M McQuillan Strategies for Implementation of Multi-Host Computer Networks ACS IFIP Joint International Symposium on Data Communications Technology and Practice August 1976 pp 26 1-26 6 also in Computer Communication Review Vol 6 No 4 October 1976 pp 19-24 F E Heart and D C Walden Communications Applications of the Pluribus Computer Conference Record of the 1976 IEEE National Telecommunications Conference Dallas Texas 29 November-1 December 1976 pp 7 1-1 to 7 1-5 t R Levin J M McQuillan and R Schantz Distributed Systems a section of New Directions for Operating Systems Report J C Browne Operating Systems Review Vol A11 Workshop No 1 January 1977 pp 14-19 J M McQuillan and D C Walden Issues in Packet Switching Network Design and the ARPANET Design Decisions submitted to Computer Networks J Davidson W Hathaway N Mimno J Postel' R Thomas and D Walden The ARPANET TELNET Protocol Its Purpose Principles Implementation and Impact on Host Operating System Design submitted for presentation at the ACM IEEE Fifth Data Communications Symposium Snowbird Utah September 27-29 1977 I I C negi e-Mllon University t NASA Ames Research Center USC Information Sciences Institute A-9 Report No B 4799 Bolt Beranek and Newman Inc Selection of ARPANET Logical Maps The following logical maps of the ARPANET are included December 1970 April 1971 August 1971 March 1972 August 1972 September 1973 June 1974 November 1974 January 1975 June 1975 October 1975 July 1976 October 1976 January 1977 March 1977 B-i IMP 360 75 9 360 67 06' MP IMP IMP onP-sls POP-IO CA IMP HARVARD UCLA IMP IMP IMP IMP mm 360m 1333 009-516 POP-IO ARPA NET DECEMBER 1970 fm mivngw1-1 0 0 0 0 IL I I I 0 Iw CiL IIL JIA I IoL 8 I I L 0 0 to OI WE 0 I IL 4L4 pm n - 3 Ir For-Io 418m PDP II 4 661611 ILL a I - 316 - 90 14-645 i ARES TIP 360 67 diffs Q ma STANFORD 0659 350 75 IMP IMP TSP ARPA UNIVAC 4mm - mac 7116' 3i 6 ucu sac vs NOAA mp c413 IMP IMP TIP TIP TIP IMP 360st 5'30 nap-me 360 44 I mam Iv IMP I 86700 0050 570 14 ARPA NETWORK LOGICAL MAP AUGUST I972 c Nlo 0 0 Z s 0 A 0 XLz 0 0 o 0 I -0_ 9_0 zI M L0 4 I- W Oc 47j 0 0 03 a A cc0 f--I 0 w p 0 ICI 1 w IL- aWc -- C0 I -l 0 g aacal a - CI L CL C IL 0a aCl a b a 4 M MrI 0 z U jf I% a U 5 0 0 ff 0 0 - 100 CL I4 0 LC 0 9L CL 0 w AL awL 0 D x 4L 40 9LL C -0 0 0 a t CL IL 0 0 cL 0 CL 0 0 cL 0 z in u v z a 0 cc4 u w CL 0 one 0 us a Q CL 9 w z 000 0000 40 a 00 t r 0 a OU -j u u U ow 2 0u 0u C% 4-1 Inc 7o13 w La w 0- 0 w w -4 1 y CL cc w CL ob I to of w 0 W a a- 0 a fn z CLAL 0 qr z UO UV a in 0 N 00 CL CL 0 CL CL 0- a IL CL w CL 0w v 1A c t CL ca X -3 0 It 0 o 0 o a U$ w 0 cr 0il z a7 0u 31 LLU 43 IL in W Z 0 4n z 0 49 C03 x w 10- ob 0 CL 8 a Mc 0 0 u 010 0 1 10 Ion F Q x u on IL u 0 3 y z x 0 AM a MA co W 00 X C I- z M u vi WS I-U U 0 z Lu 0 00to W CL CL w V o 2 w i qj x 04 x 0 -K 39 CC P10 % 0 0 Fm m z in 2 x X U 0 Z u z Z X'd 3 x CL 4L w M IL CL w IL 0IL ki IL 44 --ildraiffmifl- ICL sn a 00 'u 0 0 x GM U 440ir W OX                     National Security Archive    Suite 701  Gelman Library  The George Washington University    2130 H Street  NW  Washington  D C  20037    Phone  202 994‐7000  Fax  202 994‐7005  nsarchiv@gwu edu