[IMR] IMR88-05.TXT MAY 1988 INTERNET MONTHLY REPORTS ------------------------ The purpose of these reports is to communicate to the Internet Research Group the accomplishments, milestones reached, or problems discovered by the participating organizations. This report is for research use only, and is not for public distribution. Each organization is expected to submit a 1/2 page report on the first business day of the month describing the previous month's activities. These reports should be submitted via network mail to Ann Westine (Westine@ISI.EDU) or Karen Roubicek (Roubicek@NNSC.NSF.NET). BBN LABORATORIES AND BBN COMMUNICATIONS CORPORATION --------------------------------------------------- WIDEBAND NETWORK The three cross country Arpanet Inter-Switch Trunk (IST) lines which send traffic over the Wideband Net saw their first full month of operation in May. The performance of these lines over most of the month was less than desirable, but through the combined efforts of all parties, the Wideband Net and attached Arpanet IMPs were tuned so that the IST lines can now carry moderately heavy levels of Arpanet traffic and provide acceptable service to the network users. During the most recent test, we looped all cross country land lines which are scheduled to go away in the near future, and the Wideband IST trunks performed quite well. The traffic levels during this test, however, were unusually light, so we plan more tests during high traffic periods. Westine [Page 1]  Internet Monthly Report May 1988 Delay has been the primary problem in making the Wideband / Arpanet IST lines perform smoothly. Several additional changes to address this problem will be fielded in the Wideband Net shortly. Software to reduce delay in the BSAT host interface, to provide delay measurements, and to improve monitoring will go out shortly. Also, a modified hardware I/O interface between the Wideband BSATs and the IMPs which will reduce delay is also being tested. SATNET The SATNET has been very stable through the month of May. We did not have any unscheduled outages of the SATNET SIMP's or PSP terminal hardware. The availability of the SATNET was again above 99% from tests run by ISI for all sites but Tanum. The Tanum earth station has been preparing to repoint its dish to a new satellite and as a result has been periodically removing the SIMP from the channels. The new UCL to RSRE and RSRE to Goonhilly lines are now in place. The two new links are now 64 Kb Kilostream service and have been operational for several weeks. INTERNET R&D The Internet continues to grow. We are currently at about 435 operational networks. We have almost completed the Telenet certification of the Butterfly VAN Gateway gateway X.25 interface. So far we run and passed all of the tests except for reverse charging. We expect to complete that series of tests in June and submit the results to Telenet. We have also completed the development of the Pronet Ring interface and are in the process of debugging the code. We hosted a meeting between members of the Open Routing Working Group and representatives of common carriers. The meeting was organized by Noel Chiappa. The subject of the meeting was the Switched Multi-megabit Data Service (SMDS) and issues of the interconnection of public networks and private internets. SMDS is a user interface and a subnet requirements specification. The goal of SMDS is to provide LAN service over a Metropolitan Area Network and ultimately over a Wide Area Network. The protocol is services/marketing driven. They appear to want to drive the technology to meet the requirements. The interface to SMDS consists of two parts, the SMDS Interface Protocol (SIP) and the Subscriber network interface (SNI). The SIP Westine [Page 2]  Internet Monthly Report May 1988 contains 3 (non-iso) layers. Level 3 is the network services layer. The header uses CCITT E.164 (ISDN) addresses and has provision for Inter-Lata routing information. This is currently a Nynex and Bellcore joint project, but it is clear they mean to expand to include other carriers. The PDU can contain up to 8191 octets of user data. Level 2 provides framing and error detection. Level 1 uses a DS3 interface. The SNI offers different access classes and based on the access class, controls the flow rate, including burstiness. The algorithm used is a leaky bucket algorithm and assumes that hosts will hold back flows according to their "credit balances". Bob Hinden (Hinden@BBN.COM) ISI --- Internet Concepts Project Jon Postel hosted the CERFNET meeting at ISI, May 11th. Two RFCs were published this month. RFC 1029: Parr, G., "A More Fault Tolerant Approach to Address Resolution for A Multi-Lan System of Ethernets", May 1988. RFC 1054: Deering, S., "Host Extensions for IP Multicasting", May 1988. Ann Westine (Westine.ISI.EDU) Multimedia Conferencing Project Dave Walden completed the debug and check-out phases of the dual- mode version of PVP (the packet video software in the Butterfly machine). This allows the operator to specify via keyboard input either a Multi-Point or Point-to-Point video ST connection and thus to direct PVP to to use the corresponding messagecontrol format. This will enable us to test PVP with the initial version of the Butterfly ST gateway. Last month, an unusually high number of video screen glitches in transmissions from BBN to ISI was observed. Because the error rate was asymmetrical, a hardware problem was suspected. After testing, it was found that the errors were due to a defective board in the ESI at BBN. Westine [Page 3]  Internet Monthly Report May 1988 The conference control program which runs in a small window on the Sun screen was used this month to automate conference initiation. It currently initiates voice connections, although work is underway for video connections as well. It was expanded to supplement its local video camera control functions with remote capabilities. In addition, it was generalized to consider multisite conferencing scenarios. The eventual goal is for it to merge with the already existing mmconf program that serves as the conference initiation mechanism for Diamond multimedia documents. The conferencing system was used for a tele-meeting of the Host Requirements Working Group of the Internet Engineering Task Force on May 18th. The all-day meeting was conducted without headphones and relied upon the performance of an echo cancellor. The sound quality of this set up received mixed reviews which has warranted further investigation. Brian Hung began the first phase of the echo canceller test. This includes the simple feed through of near-end serial mu-law data to parallel mu-law output port, plus the conversion of near-end serial mu-law data to linear and from linear back to mu-law before writing the data to the parallel output port. Currently Brian is determining the echo response from our 4-to-2 wire hybrid due to far-end signals. The determination of this will enable him to devise a near-end speech detection mechanism to inhibit the updating of the tap weights while a near-end signal is present. Steve Casner, Dave Walden, Eve Schooler, Brian Hung (casner@ISI.EDU, djwalden@ISI.EDU, schooler@isi.edu, hung@ISI.EDU) NSFNET Project Annette DeSchon and Walt Prue attended the TCP Performance Seminar in Monterey, Ca. on May 9-10. They presented this material to an ISI audience on May 19th. Annette DeSchon continued work on Background File Transfer (BFTP). BFTP is now able to resume the transfer of multiple files in the middle of the list, following a loss of network connectivity or the crash of either the source or the destination host. Bob Braden served as chair of the IETF Working Group on Host Requirements, May 18th, and worked on editing the draft document the group is putting together. The current draft is 80+ pages, and incorporates text written by 6 people. The group held a one-day video teleconference between ISI and BBN to discuss the document; Westine [Page 4]  Internet Monthly Report May 1988 12 people attended. The decisions from that meeting have been incorporated into the document. Bob Braden and Annette DeSchon (Braden@ISI.EDU, DeSchon@ISI.EDU) Supercomputer and Workstation Communication Project Alan Katz continued work on a split editor under GNU Emacs written in Emacs Lisp. Alan Katz (Katz@ISI.EDU) MIT-LCS ------- No internet related progress to report. Lixia Zhang (Lixia@XX.LCS.MIT.EDU) NTA-RE and NDRE --------------- No report received. SRI --- No Internet related progress to report. Zaw-Sing Su (zsu@tsca.istc.sri.com) UCL --- Infrastructure: Kirstein and Crowcroft attended the SATNET and Infrastructure and ICB meetings at FGAN, Bonn, West Germany. A number of important issues concerning multi-agency International Network Interconnection were discussed. Some members of the FRICC were present, and contributed to a lively discussion on policies for interconnection and routing. The possible replacements for SATNET, which will be decommissioned fairly soon, were also discussed at great length. Watch this space for developments. Westine [Page 5]  Internet Monthly Report May 1988 Reports on a wide range of experiments carried on over the existence of SATNET should become availble fairly soon. Research: A modified implementation of the IEEE 802.1 addendum D draft MAC level routing algorithm is now complete, and undergoing extensive tests for robustness. The algorithm seems to be reasonably stable under node outages and recovery. A number of minor errors were uncovered in the version of the draft standard we were working from. John Crowcroft (jon@CS.UCL.AC.UK) UNIVERSITY OF DELAWARE ---------------------- 1. Jeff Simpson is studying hierarchial routing algorithms with the intent of applying these to an interdomain routing protocol. The goal of this project is to improve the use of connectivity over existing hierarchial algorithms, while maintaining reliability, robustness, and autonomy of individual systems. 2. Paul Schragger devised a simple network simulator to examine two proposed transport-level congestion control/avoidance schemes. The simulation has shown that either scheme alone is stable, but the two implemented together appear to become unstable in a number of cases. This behavior has instigated research into providing a mathematical model of network communications based on digital feedback-control concepts providing a uniform analytical approach to predict the cases of flow instablity. 3. A concerted effort was made to resolve continuing performance problems between NSFNET and ARPANET hosts. Through the kind cooperation of the six NSFNET Backbone site operators, current configuration files were collected and stashed in a directory at a mutually trusted place. This exercise alone turned up a number of inconsistencies, most of which were resolved and fixed. The configurations were checked against the current routing information kept by the Backbone fuzzballs and the Internet Operations Center and other discrepancies were found. The most serious problems appear to be failure to switch the default net when the ARPANET path breaks (UIUC) and excessive route flapping (NCAR). Westine [Page 6]  Internet Monthly Report May 1988 4. Previously reported problems with excessive route and metric flaps in EGP updates from the core system persist. Some of these can be explained by the fact not all EGP core clients patter with all three EGP core servers; however, interesting anomalies have been observed where routes have been "captured" by one client when another client was squawking a lower distance, which clearly implies something is badly broken somewhere. Further quest was temporarily suspended pending the end-semester rush. 5. The Network Time Protocol (NTP) community continues to grow, with new servers online at MIT and planned for DECWRL and NASA/AMES. Spring thunderstorms glitched some clocks and occasional circuit outages affected others; however, there were no known cases where these events caused timewarps in hosts complying with the peerage rules suggested in the documentation. 6. Effective 5 July, the University of Delaware is to be disconnected from the ARPANET, but will continue service via SURANET. Our campus administration has agreed to upgrade the SURANET access line to T1 speed and to expand the existing 3- node 80-Mbps fiber ring to over a dozen nodes. Our frisky fuzzballs will remain connected to the ARPANET and everything else in town, but will not carry regular campus traffic. 7. Dr. Mills attended a meeting of the National Security Telecommunications Committee at NAS Headquarters in Washington, DC. A report covering issues in reliable telecommunication network synchronization is now in preparation. Dave Mills (Mills@UDEL.EDU) Westine [Page 7]  Internet Monthly Report May 1988 NSF NETWORKING -------------- NSF NETWORKING UCAR/BBN LABS NNSC Since the Domain Tutorial in April, the NNSC has been responding to requests to perform backup domain service for NSFNET sites. Craig Partridge chaired the second meeting of the MIB Working Group of the Internet Engineering Task Force. The group released a proposed Internet standard for the Structure of Management Information and the MIB itself. These are now available as IETF IDEAS on sri-nic.arpa. The comment period on them ends June 17th (please direct all comments to craig@nnsc.nsf.net). In Ann Arbor, Craig attended a meeting with Merit, IBM and Guy Almes of Sesquinet to discuss issues related to network management on the new NSFNET backbone. Karen Roubicek spoke about NNSC activities at the biweekly NSFNET Managers' Meeting and met with the NSFNET backbone Information Services team to discuss providing complementary information services to the NSFNET community. by Karen Roubicek (roubicek@nnsc.nsf.net) NSFNET BACKBONE PROJECT CORNELL UNIVERSITY THEORY CENTER The version of gated that had previously been available as a beta release is now the offical release. This release mainly fixes and enhances gated's EGP processing. The majority of the gated work for support of the new backbone is done, including the documentation. Gated and the EGP code in the NSSs is undergoing testing at Merit and IBM with some further testing to be done at Cornell starting in a few days. The growth in traffic of NSFNet, as indicated by the following traffic totals, are most significant considering the problems NSFNet has experienced during this six month period. The complete traffic reports are available from Doug Elias at Cornell. Westine [Page 8]  Internet Monthly Report May 1988 NSFNET TRAFFIC DATA November 1987 - April 1988 Nov '87 Dec '87 Jan'88 Feb'88 Mar'88 Apr'88 Total 333,508,417 421,008,710 558,716,009 Packets Transmitted 336,757,929 444,159,631 615,306,546 ___________________________________________________________________ Site Traffic: Percentage of Total Nov'87 Dec'87 Jan'88 Feb'88 Mar'88 Apr'88 Cornell % 22.87 23.08 24.27 21.78 22.00 20.62 JvNC % 21.62 21.74 24.40 19.12 21.80 20.03 NCAR % 18.52 19.37 19.28 19.16 17.91 16.88 UIUC % 17.72 13.86 12.98 15.44 14.34 14.93 PSC % 15.11 16.59 15.37 16.13 15.28 18.88 SDSC % 4.16 5.37 3.69 8.38 8.67 8.65 by Martyne M. Hallgren (martyne@tcgould.tn.cornell.edu) NEW NSFNET BACKBONE The NSFNET backbone project remains on schedule. Seven of the Nodal Switching Subsystems (NSSs) are installed and online as of May 31. These NSFNET sites are Ann Arbor, Boulder, Ithaca, Lincoln, Palo Alto, Princeton, and San Diego. The other six are on the test network at Yorktown Heights (IBM), Reston (MCI), Milford (IBM), and Ann Arbor (Merit). The two network partitions are linked in Ann Arbor. Over the next week, the remainder of the backbone sites are scheduled to come online. By June 10, all 19 NSSs should be installed and exchanging IS-IS routing packets. Central to full connectivity for the NSFNET backbone is the software developed for the NSSs. For routing within the NSFNET core, the NSS code incorporates the IBM adapted ANSI IS-IS Intra- Domain Routing Exchange Protocol, which uses a Shortest Path First (SPF) algorithm for internal routing. These IS-IS packets are now being exchanged between the operational NSSs. Westine [Page 9]  Internet Monthly Report May 1988 Currently, gateways at Cornell (Unix), SDCS (VMS), NCAR (Unix), and JVNC (Cisco) are running the Cornell Gateway Daemon (gated) developed at Cornell University. In addition, initial tests were done using Proteon gateways. The next step in phasing in the "new" network is to improve the routing interface between the RCPs and other local gateways via the External Gateway Protocol (EGP). This testing is being done as more mid-level networks are in the process of adding Ethernet connections between the NSSs and their local networks. In addition to NSS development, the network management system for the new backbone is being readied for operation. Netview and Netview/PC have been installed and configured at Merit's Network Operations Center (NOC) in Ann Arbor. Testing is proceeding on Simple Gateway Monitoring Protocol (SGMP) which will be used to monitor network traffic, routing tables, and link status on the backbone. SGMP testing is being done on the experimental network by researchers at Yorktown Heights with the assistance of Jeff Case and Ken Key of the University of Tennessee. File Transfer Protocol (FTP) is now available on the IBM 4381/Q14 Information Services machine known as NIC.MERIT.EDU (35.1.1.48). FTP allows users to transfer both ASCII and binary files to and from foreign hosts. Files may now be retrieved from the Merit/NSFNET IS machine via anonymous FTP. NSFNET BACKBONE SITES & MID-LEVEL NETWORK SITES UNIVERSITY OF ILLINOIS AT URBANA-CHAMPAIGN/NCSANET No report received. JOHN VON NEUMANN NATIONAL SUPERCOMPUTER CENTER No report received. NATIONAL CENTER FOR ATMOSPHERIC RESEARCH AND UNIVERSITY SATELLITE NETWORK PROJECT New radios are to be installed at all the USAN satellite sites and at the NCAR hub in mid-June. While this will improve the performance in bad weather conditions, it will necessitate that USAN be down for about four days. Westine [Page 10]  Internet Monthly Report May 1988 The NSF node at NCAR on the new NSFNET backbone was the first one to be operational and be tested. At this point it is directly attached to the USAN ethernet backbone, but plans are to isolate it on a stub ethernet with a cisco gateway to the USAN AS. by Don Morris (morris@windom.ucar.edu) PITTSBURGH SUPERCOMPUTING CENTER No report received. SAN DIEGO SUPERCOMPUTER CENTER Our ARPANET PSN is online and operational. Our gateway machine, a MicroVAX II using an ACC5250 interface, is 10.1.0.26 (lola.sdsc.edu). The line to UCLA is still not terminated in the remote PSN. The new NSFnet NSS has been assembled at SDSC. It is now exchanging packets with other parts of the new net. We are exchanging EGP information with it. The p4200 continues without any anomallies. We hope to soon upgrade it with Proteon's newer CPU, Ethernet cards (a pair), and level 8.0 software. The University of Hawaii converted their SDSCnet (MFEnet protocol) line to a p4200 link. Currently our single p4200 is gatewaying BARRnet(56), Hawaii(56), NWNet(56), Salk Inst.(T1,IP&DECnet), SDSC(E-net), UCSB(56), and UCSD(2nd E-net). We attempted to move our GATED functions from our small SUN (3/50) to one of the VAX's during May. The SRI MultiNet software was very unstable when used in the active mode. The SUN is again carrying the load. by Paul Love (loveep@sds.sdsc.edu) BARRNET (January - April) This report briefly covers BARRNet actiivity since January, at which time we were not yet exchanging routing information with the NSFNET backbone, except for selected supercomputer destinations. In February-March, parallel efforts at SDSC and within BARRNet to permit advertising NSFNet routes into and out-of BARRNet without breaking local routing and reachability elsewhere culminated in Westine [Page 11]  Internet Monthly Report May 1988 activation of BARRNet as a "real" regional NSF network. However, various routing and performance muck in the easterly networking domains delayed activation of BARRNet route exchange for a while to avoid making matters worse. By the end of March, however, BARRNet sites were exchanging routing with the other regionals and supercomputer networks and general BARRNet/NSFNet connectivity was achieved and being used within the constraints imposed within the individual member networks. Stanford, for example, has limited NSFNet routing exchange to selected remote supercomputer networks because its internal subnet routing systems must be substantially modified to permit generalized external internet route selectivity from within campus. Also in March, the first BARRNet Annual Report was submitted to the NSF, covering the period from award of grant to January 31, 1988. Copies are available on request from gd.why@forsythe.stanford.edu. During March and April, technical planning has focused on preparation for the new NSFNET backbone connection. The BARRNet Technical Committee has decided on a BARRNet/NSFNET NSS connection configuration consisting of: (a) a subnetted LAN to which the NSS and SU BARRNet gateways attach (must be a single LAN for both with only 1 network number in use on the wire). (b) the NSS is told the LAN is NOT subnetted. (c) the NSS is told explicitly what broadcast address to use on the LAN (which of course is the subnet IP broadcast address). (d) the NSS runs EGP but not RIP. (e) any SU campus gateways attached to this net will need to do proxy ARPs for SU hosts. BARRNet will obtain its own AS number. Planning for additional member sites is actively underway....the first of which is expected to be LLNL (Lawrence Livermore Labs), hopefully in June. Plans for upgrade of BARRNet backbone facilities to higher performance routers to support membership expansion have also been approved by the technical committee and consortium members. This will be achieved by substitution of existing router components into stub networks as new members are added. BARRNet representatives also participated in the Federation and Net'88 meetings. by Bill Yundt (GD.WHY@forsythe.stanford.edu) MERIT/UMNET A T1 line from Michigan State University to the University of Michigan has been installed and is now functional. Most Merit links Westine [Page 12]  Internet Monthly Report May 1988 are being upgraded to at least 56 Kbps with T1 links installed in certain places. We are also now implementing SLIP (Serial Line Internet Protocol) which will be accessible from any hardwired Merit port. by Laura Kelleher (laura_kelleher@um.cc.umich.edu) MIDNET No report received. MRNET During May, NSFnet appeared to be more reliable than during April, but it is still not without problems. Some users report difficulties regularily, while others seem to be able to use NSFnet without much difficulty. Because traffic can be routed either through NSFnet, or through ARPAnet, it is often difficult to determine the real source of problems that are reported. One ethernet controller on the Proteon router failed during May, and is currently being repaired. The MRNet technical committee is going to try to more clearly identify the problems being encountered with NSFnet; hopefully more data will be available for our report next month. by Ken Carlson (kgc@uf.msc.umn.edu) NORTHWESTNET No report received. NYSERNET NYSERNet had the following topology: Westine [Page 13]  Internet Monthly Report May 1988 Clarkson NYSERNet/HQ | | | | Rochester=====Cornell==Syracuse===RPI---Albany || || || || ....Alfred || || Buffalo...Fredonia || || | ....Oswego || || | || || Binghamton || +-------- || ------StonyBrook | || | || | || | || CUNY--NYTEL/NSMAC--Columbia======NYU==Rockefeller | |\ | | /| | || | | \ | NYNEX/S&T / | | || | | \ BNL / | | WP/CO | | \ / | | | | +-------------POLY---+ | | | | | | | +------------------------------+ | | | +--------------------------------------+ || ==== || T1 || \ | / 56kbits .... 9.6kbits In May: 1) The Syracuse/RPI link was upgraded to T1 2) The Rochester/Cornell link was upgraded to T1 3) the Buffalo/Rochester link was upgraded to T1 4) Mark Fedor participated in the unified network management Management Information Base (MIB) working group of the IETF. 5) Marty Schoffstall and Wengyik Yeong participated in the Simple Network Management Protocol (SNMP) working group. 6) A prototype of the new SNMP based on the draft documents of the working group was completed. by Martin Schoffstall (schoff@nisc.nyser.net) Westine [Page 14]  Internet Monthly Report May 1988 OARNET No report received. SESQUINET The complete initially proposed SesquiNet configuration, now augmented by Prairie View A&M, has been operational now for several months. The following campus networks are being served, and are advertised via EGP to the core: Baylor College of Medicine 128.249 BCM-Technologies 192.31.88 Houston Area Research Center 192.31.87 Prairie View A&M University 129.208 Rice University 128.42 Texas A&M University 128.194 Texas Southern University 192.31.101 and the University of Houston 129.7 The serial line from NSFnet/NCAR to SesquiNet/Rice has been operational for several months, and routes to SesquiNet via NSFnet are now being advertised. Performance is very good. Our smoothest Arpanet experience came during the early part of May when our cisco Arpanet support had been debugged and our IMP and lines were performing well. During the middle of the month, however, we were disconnected from our IMP, probably never to return. For a week we used only the interim NSFnet for external connectivity; this gave us good connections to many sites and inadequate connections to others. In late May, through the courtesy of Merit, our nets were advertised by the Arpanet gateway at Merit, and our connectivity to the Arpanet was radically improved. Our new NSFnet Backbone gateway is being assembled as I write, and we will working this month to bring it into operational use within Sesquinet. We will soon resume our testing cisco's support for dual protocol (IP and DECnet) routing. by Guy Almes (almes@rice.edu) SURANET No report received. Westine [Page 15]  Internet Monthly Report May 1988 WESTNET 1. The NSF FY'88 award arrived late in May. Pursuant to this, we have initiated the paperwork to pay a number of outstanding bills. 2. We are in the process of setting up interviews for the Westnet Technical Support Position. We presently have narrowed the field down to three excellent candidates. Interviews will commence on June 9, 1988. 3. David C. M. Wood of the University of Colorado at Boulder has been testing with Guy Almes of Rice University cisco's implementation of EGP. To date, everything appears to be working well. 4. The installation of the IBM NSS at the University of Utah is proceeding. Estimated power-up is Friday, May 27, 1988. The cisco to be connected to the NSS via an Ethernet is apparently an optimal configuration for safe testing as it has only one Ethernet connected only to the NSS with a Class C address. Brigham Young University has volunteered to serve as an initial test site. 5. We have, for the first time, collected some statistics (albeit primitive) from the Westnet cisco Gateways. Results indicate that the circuit between NCAR and the University of Colorado at Boulder experiences very heavy use, and the indication is that it is saturated for a good portion of the eight hour day. We are investigating upgrading this circuit to T-1 speed. 6. We are vigorously into the planning stage for the Westnet Steering Committee meeting, to be held in Denver on Thursday, July 7, 1988. Issue to be addressed include: traffic, status of the IBM NSS node at the University of Utah, campus network ing activity, staffing, training, and preparation of the NSF renewal proposal. Patrick J. Burns (pburns@csupwb.colostate.edu) Westine [Page 16]  Internet Monthly Report May 1988 TASK FORCE REPORTS ------------------ APPLICATIONS -- USER INTERFACE The Video Working Group, chaired by Steve Casner, held a meeting in Boston on May 10. Institutions / projects represented included Olivetti Research, the MIT Media Lab, David Sarnoff Research Center, USC/ISI, MIT Project Athena, BBN Labs, Parallax Graphics, Sun Microsystems, Xerox PARC, Bellcore, and the Columbia University Center for Telecommunications Research. A report of that meeting is forthcoming. The next meeting of the Task Force is scheduled for June 22-23, at Olivetti Research Center, in Menlo Park. Keith Lantz (Lantz@SCORE.STANFORD.EDU) AUTONOMOUS NETWORKS The Autonomous Networks Task Force will hold a meeting July 7 and 8, using the teleconferencing facilities at BBN and ISI. The meeting is intended as a follow up to a June meeting with FRICC representatives at NASA/RIACS. Deborah Estrin (Estrin@OBERON.USC.EDU) END-TO-END SERVICES No internet related progress to report. Bob Braden (Braden@ISI.EDU) INTERNET ARCHITECTURE Activity was low this month. The next meeting is to be called after the dust settles from Barry Leiner's meeting at RIACS in late June. Dave Mills (Mills@HUEY.UDEL.EDU) INTERNET ENGINEERING The preliminary agenda for the June 15-17 IETF meeting is below. The basic format remains the same as the last several meetings, namely, Westine [Page 17]  Internet Monthly Report May 1988 1) Working Group meetings for the first 1.5 days, 2) Presentations and network reports on Thursday afternoon and 3) Detailed working group reports, technical presentations and discussion on Friday. The final agenda, including the Working Group schedule, will be in next month's report. Proceedings of the March IETF will be distributed at the June 15 meeting and will be available from the NIC. Agenda for the June 15-17 IETF ------ --- --- ---- ----- ---- WEDNESDAY, June 15 9:00 am Opening Plenary (Introductions and local arrangements) 9:30 am Working Groups (schedule to be announced) 5:00 pm Recess THURSDAY, June 16 9:00 am Opening Plenary 9:15 am Working Groups (schedule to be announced) 1:00 pm Network Status Reports o Arpanet/Internet Report (Hinden/Lepp (Gardner), BBN) o Status Report on the New NSFnet (Braun, UMich/Rekhter, IBM) o FRICC Initiatives (Wolff, NSF/Bostwick, DOE) o Canadian Research Networking (Curley, NRC of Canada) o Switched Multi-Megabit Data Service (SMDS) (Singh, NYNEX) 5:00 pm Recess FRIDAY, June 17 9:00 am Working Group Reports and Discussion 1:00 pm Technical Presentations o TCP Performance and Other Unconfirmed Rumors (Van Jacobson, LBL) o Cray TCP Performance, Update (Borman, Cray) o Bellringing and Gongferming (Mills, UDel) o Issues in Canadian Networking (Prindeville, McGill) Westine [Page 18]  Internet Monthly Report May 1988 4:45 pm Concluding Plenary Remarks 5:00 pm Adjourn Phil Gross (gross@gateway.mitre.org) INTERNET MANAGEMENT No report received. PRIVACY Preparations continued for the next Privacy Task Force meeting, to be held on 15-16 June 1988, at DEC, Littleton, MA, hosted by Morrie Gasser. The Privacy Task Force welcomed a new member, Russ Housley of Xerox Special Information Systems. John Linn (Linn@CCY.BBN.COM) ROBUSTNESS AND SURVIVABILITY No report received. SCIENTIFIC COMPUTING No report received. TACTICAL INTERNET No report received. TESTING AND EVALUATION No report received.