Transport Area Director: o Allison Mankin: mankin@cmf.nrl.navy.mil Area Summary reported by Allison Mankin/NRL The Transport Services Area deals with protocols and algorithms that provide end-to-end transmission services in the Internet. We maintain the notion of transport services, not just transport protocols, because of the increasing variety of end-to-end requirements that the Internet meets or will have to meet soon. The members of the Transport Area Directorate are: Dave Borman dab@cray.com Sally Floyd floyd@ee.lbl.gov Jim Hughes hughes@hughes.network.com Matt Mathis mathis@pele.psc.edu Greg Minshall minshall@wc.novell.com Eve Schooler schooler@isi.edu John Wroclawski jtw@lcs.mit.edu Lixia Zhang lixia@parc.xerox.com The Transport Area Directorate is primarily responsible for quality review of the working groups. It meets regularly on the MBONE to plan for each upcoming IETF. It has a mailing list, but its location is in transition, with the Area Director's move to ISI (in a new East Coast lab). Here is some news of the Area from between the meetings and San Jose, followed by group meeting summaries. o We will soon plan what Transport Next Generation work should begin, as the mid-term result of the TCPng BOF reported below. o The ONCRPC Working Group was in hiatus for San Jose, while the IETF's agreement with Sun was worked out. Soon on the TSV agenda will be determining how the second half of the ONCRPC charter (enhancing the standards once they were documented) should go. o Richard Stevens, the author of TCP/IP Illustrated, has been asked to write an RFC documenting TCP's adaptive algorithms, since Sally Floyd reviewed that chapter in TCP/IP Illustrated Vol. 2 and reported that it was extremely accurate and clear. He will try to do so before the next IETF. o The Thin-OSI Working Group completed its charter and concluded between Toronto and San Jose. o The TMUX proposal, developed in a TSV BOF, received an extended Last Call and was elevated to Proposed Standard (RFC 1692). o Of the TSV working groups, four met in San Jose, and one BOF was hosted, TCPng, a review in conjunction with IPng. o The ONC Remote Procedure Call Working Group (ONCRPC) and the TCP Large Windows Working Group (TCPLW) are dormant. Next Generation TCP BOF (TCPNG) The TCPng BOF was convened to consider what changes need to be made to TCP for the transition to IPv6, and whether any other new features should be added at this time. The meeting distinguished TCP6, the minimal changes to operate over IPv6, from a TCPng, a wonderful new TCP-like protocol. It discussed a number of different possible features that might be included in TCPng, and suggested that the Transport Service Area Director might set up a working group to talk about next generation transport protocols. Audio/Video Transport Working Group (AVT) A report was presented on the changes to the Real-time Transport Protocol draft since the July version (-05). These changes were all small, though a few did introduce incompatibilities. There were no objections to these changes. However, there was a surprising amount of debate about the jitter parameter in the Reception Report which had not changed except that the algorithm had been defined. As a result, a second session was scheduled to allow completing the planned presentations. During the second session, a compromise solution was devised: the jitter field remained unchanged, but the ``cumulative packets received'' was changed to ``cumulative packets lost'' and reduced to 24 bits. The freed 8 bits will contain a ``fraction of packets lost'' allowing short-term loss measurements with singleton reception reports. With this issue settled, the draft editing will be completed and the draft will be submitted for Area Directorate review. Additional presentations were given on the implementation of RTPv2 in vic (by Steve McCanne) and Loki (by Frank Kastenholz), and on the new draft Packet Format for Encapsulation of MPEG in RTP (by Gerard Fernando). Integrated Services Working Group (INTSERV) The integrated services working group (INT-SRV) held two meetings at the San Jose IETF. At these meetings the group discussed several parts of a ``reusable framework'' approach to providing quality of service control in the Internet. The key goal of this approach is to allow new services to be deployed in an evolutionary, market-driven manner, while maintaining backward compatibility and (re)using standard components. The first meeting began with a brief presentation describing this new approach and how it differs from the group's previously proposed path. John Wroclawski's presentation discussed goals, described four components (QM manager, setup protocols, traffic specifiers, and service specification templates), and briefly described the impact of this approach on the work items adopted at the Toronto IETF. The majority of the first meeting was devoted to a discussion of the Quality of Service Manager, an abstraction layer designed to isolate applications from the specific details of services provided by an internet. Dave Clark gave a general presentation of the QM idea, its uses, and how it might be realised in practice. Craig Partridge discussed a binding of the QM to Berkeley sockets. Both of these presentations describe early work, and the authors received significant feedback from the working group. The meeting closed with a brief discussion of the merits of the framework approach to providing integrated services. A show of hands revealed essentially complete consensus that the new approach was preferred to the alternative of mandating a single set of services throughout the Internet. The group continued the discussion of framework components in the second meeting. Scott Shenker presented a preliminary draft of a service specification requirements document. The document has two parts. The first specifies a ``template'' for defining services, including router behavior, composition rules, and the like. The remainder of the document gives several examples of the use of this template. The intention is that the template format become a standard, but for now, the example services are informational. The group agreed that this draft was a good base for continued work. Craig Partridge gave a presentation about traffic specifications (``flowspecs'') and whether these are a reusable component or specific to a particular service. After some discussion, the group appeared to find some merit to common flowspecs, but believed that further concrete evidence of their usefulness was required. Further discussion centered on the use of RFC 1363 or the Windows Sockets version 2.0 specification as a possible starting point for experimental implementations. John Wroclawski presented a proposal for initial deployment of integrated service capabilities in the July '95 time frame. The proposal includes RSVP, a level-of-effort predictive service, and management and monitoring capability. John asked about interest in a small implementors group to further this work, and received several positive responses during and after the meeting. The second meeting closed with a technical discussion about token bucket filters. TBFs are widely proposed today as traffic descriptors and shapers, but recent work by Mark Garrett and Craig Partridge suggests some limitations. This is directly relevant to the flowspec discussion described above, and will need to be better understood in the future. Multiparty Multimedia Session Control Working Group (MMUSIC) In an effort to make the working group aware of other standardization efforts, two brief presentations were given on ITU teleconferencing efforts. Next, an overview of both a centralized and a distributed implementation of the Shenker-Weinrib-Schooler agreement protocol was given -- the implementation experience gave rise to several interesting research issues concerning generality, dynamicity, and scalability. A summary of the responses from a survey of existing session control protocols was then given, focusing on derived functional requirements for both session control protocols and session descriptions. An overview of inter-system (horizontal) and intra-system (vertical) session control requirements was presented which left open questions on the transport and semantic requirements. A presentation was then given which identified additional requirements for the ``sd'' protocol and a discussion of proposed extension followed. Finally, the current goals and milestones were reviewed and a list of potential working group documents was discussed. Resource Reservation Setup Protocol Working Group (RSVP) The RSVP Working Group met for two sessions during the San Jose IETF. The first session began with a brief review of implementation status and a description of the MM '94 demo. The rest of the first meeting was devoted to reports from all those who took study action items at the Toronto meeting. Each report included a recommendation to the working group. The second session began with reviews of some new technical issues, not previously discussed. Then the working group developed a list of features to be included in version 1 RSVP. This list will serve as the basis for revision of the specification and review, hopefully allowing submission as a Proposed Standard in July 1995.