ABSTRACT
"Soft state" is an often cited yet vague concept in network protocol design in which two or more network entities intercommunicate in a loosely coupled, often anonymous fashion. Researchers often define this concept operationally (if at all) rather than analytically: a source of soft state transmits periodic "refresh messages" over a (lossy) communication channel to one or more receivers that maintain a copy of that state, which in turn "expires" if the periodic updates cease. Though a number of crucial Internet protocol building blocks are rooted in soft state-based designs --- e.g., RSVP refresh messages, PIM membership updates, various routing protocol updates, RTCP control messages, directory services like SAP, and so forth --- controversy is building as to whether the performance overhead of soft state refresh messages justify their qualitative benefit of enhanced system "robustness". We believe that this controversy has risen not from fundamental performance tradeoffs but rather from our lack of a comprehensive understanding of soft state. To better understand these tradeoffs, we propose herein a formal model for soft state communication based on a probabilistic delivery model with relaxed reliability. Using this model, we conduct queueing analysis and simulation to characterize the data consistency and performance tradeoffs under a range of workloads and network loss rates. We then extend our model with feedback and show, through simulation, that adding feedback dramatically improves data consistency (by up to 55%) without increasing network resource consumption. Our model not only provides a foundation for understanding soft state, but also induces a new fundamental transport protocol based on probabilistic delivery. Toward this end, we sketch our design of the "Soft State Transport Protocol" (SSTP), which enjoys the robustness of soft state while retaining the performance benefit of hard state protocols like TCP through its judicious use of feedback.
- 1.AMIR, E., MCCANNE, S., AND KATZ, R. Receiver-driven Bandwidth Adaptation for Light-weight Sessions. In Proceedings of A CM Multimedia '97 (Nov. 1997), ACM.]] Google ScholarDigital Library
- 2.AMm, E., MCCANNE, S., AND KATZ, R. An Active Service Framework and its Application to Ileal-time Multimedia Transcoding. In Proceedings of SIGCOMM 1998 (Vancouver, Canada, Sep 1998), ACM.]] Google ScholarDigital Library
- 3.BALAKRXSHNAN, H., RAHUL, H. S., AND SESHAN, S. An Integrated Congestion Management Architecture for Internet Hosts. In Proceedings of SIGCOMM 1999 (Cambridge, MA, Sep 1999), ACM.]] Google ScholarDigital Library
- 4.BANERJEA, A., FERRARI, D., MAH, B., MORAN, M., VERMA, D., AND ZHANC, H. The Tenet Ileal-Time Protocol Suite: Design, Implementation, and Experiences. IEEE/ACM Transactions on Networking (1995).]] Google ScholarDigital Library
- 5.BASKETT, F., CHANDY, M., MUNTZ, R., AND PALAGIOS, F. Open, Closed, and Mixed Networks of Queues with Different Classes of Customers. Journal of the Association for Computing Machinery ~2, 2 (1975), 248-260.]] Google ScholarDigital Library
- 6.BiitMAN, K., CHIPER, A., AND STEPHENSON, P. Lightweight Causal and Atomic Group Multicast. ACM 7kansactions on Computer Systems 9, 3 (Aug. 1991), 272-314.]] Google ScholarDigital Library
- 7.BmMAN, K., IIAYDEN, M., OZKASAP, O., XIAO, Z., BuoiV, M., ANO MINSKY, Y. Bimodal Multicast. Tech. Rep. TR98- 1683, Cornell University, Ithaca, NY, May 1998.]] Google ScholarDigital Library
- 8.BOLOT, J.-C., AND TURLETTI, T. A Rate Control Mechanism for Packet Video in the Internet. In Proceedings IEEE lnfocom '94 (Toronto, Canada, June 1994), ACM.]]Google Scholar
- 9.BOLOT, J.-C., TURLETTI, T., AND WAKEMAN, I. Scalable Feedback Control for Multicast Video Distribution in the Internet. In Proceedings of SIGCOMM '94 (University College London, London, U.K., Sept. 1994), ACM.]] Google ScholarDigital Library
- 10.CHANDY, K. M., RIFKIN, A., AND SCHOOL~n, E. Using Announce-Listen with Global Events to Develop Distributed Control Systems. Concurrency: Practice and Experience (1998), 1021-1027.]]Google Scholar
- 11.CHESSON, G. XTP/Protocol Engine Design. In Proceedings of the IFIP WG6.1/6.~ Workshop (Riischlikon, May 1989).]]Google Scholar
- 12.ChAr, g, D. D. The Design Philosophy of the DARPA Internet Protocols. In Proceedings of SIGCOMM '88 (Stanford, CA, Aug. 1988), ACM.]] Google ScholarDigital Library
- 13.CLANK, D. D., AND TENNENHOUSE, D. L. Architectural Considerations for a New Generation of Protocols. In Proceedings of SIGCOMM '90 (Philadelphia, PA, Sept. 1990), ACM.]] Google ScholarDigital Library
- 14.COVER, T. M., AND THOMAS, J. A. Elements of Information Theory. John Wiley and Sons, Inc., 1991.]] Google ScholarDigital Library
- 15.DEERING, S., I~S'I'RIN, D., FAItINACCI, D., AND JACOBSON, V. An Architecture for Wide-Area Multicast Routing. In Proceedings of SIGCOMM '94 (University College London, London, U.K., Sept. 1994), ACM.]] Google ScholarDigital Library
- 16.DELGROS$I, L., HERRTWICH, P~. G.,, HOFFMANN, F. O., AND SCHALLER, S. Receiver-initiated Communication with ST-II. Multimedia Systems 2, 4 (Oct. 1994), 141-149.]] Google ScholarDigital Library
- 17.DEMERS, A., KESHAV, S., AND SHENKER, S. Analysis and Simulation of a Fair Queueing Algorithm. In Proceedings of SIGCOMM '89 (Sept. 1989), ACM.]] Google ScholarDigital Library
- 18.FAI,K, A., AND PAXSON, V. Minutes of the "RUTS" IETF BOF, Dec. 1998. ftp://ftp.ee.lb!.gov/ietf/ruts-98-minutes.]]Google Scholar
- 19.FLOYD, S., AND JACOBSON, V. Link-Sharing and Resource Management Models for Packet Networks. IEEE/ACM Transactions on Networking 3, 4 (Aug. 1995), 365-386.]] Google ScholarDigital Library
- 20.FLOYD, S., JACOBSON, V., MCCANNE, S., LIU, C.-G., AND ZHANG, L. A Reliable Multicast Framework for Light-weight Sessions and Application Level Framing. In Proceedings of SIGCOMM '95 (Boston, MA, Sept. 1995), ACM.]] Google ScholarDigital Library
- 21.HANOLEY, M. SAP: Session Announcement Protocol. Internet Draft, Nov 19, 1996.]]Google Scholar
- 22.IIANDLr. Y, M. Session Directories and Internet Multicast Address Allocation. In Proceedings of SIGCOMM 1998 (Vancouver, Canada, Sep 1998), ACM.]] Google ScholarDigital Library
- 23.HANDI,EY, M., AND CROWCROFT, J. Network Text Editor (NTE): A Scalable Shared Text Editor for the MBone. In Proceedings of SIGCOMM 1997 (Cannes, France, Sep 1997), ACM.]] Google ScholarDigital Library
- 24.HANDLEY, M., AND JACOBSON, V. sdr- A Multicast Session Directory. University College London.]]Google Scholar
- 25.HEORICK, C. Routing Information Protocol. Rutgers University, June 1988. RFC-1058.]] Google ScholarDigital Library
- 26.JACOBSON, V. Session Directory. Lawrence Berkeley Laboratory, ftp://ftp.ee.lbl.gov/conferencing/sd.]]Google Scholar
- 27.JACOBSON, V. Congestion Avoidance and Control. In Proceedings of S~GCOMM '88 (Stanford, CA, Aug. 1988).]] Google ScholarDigital Library
- 28.JACOBSON, V. SIGCOMM '94 Tutorial: Multimedia conferencing on the Internet, Aug. 1994.]]Google Scholar
- 29.JACOBSON, V., AND MCCANNE, S. LBL Whiteboard. Lawrence Berkeley Laboratory ftp://ftp.ee.lbl.gov/conferencing/wb.]]Google Scholar
- 30.JACOBSON, V., AND MCCANNE, S. Visual Audio Tool. Lawrence Berkeley Laboratory. ftp://ftp.ee.lbl.gov/conferencing/vat.]]Google Scholar
- 31.JAIN, R. Congestion Control in Computer Networks: Issues and Trends. IEEE Network Magazine (May 1990), 24-30.]]Google Scholar
- 32.JAIN, R., P~AMAKRISHNAN, K., AND CH{U, D.-M. Congestion Avoidance in Computer Networks With a Connectionless Network Layer. Tech. Rep. DEC-TR-506, Digital Equipment Corporation, Aug. 1987.]]Google Scholar
- 33.LOUGHEED, K., AND REKHTER, Y. A Border Gateway Protocol (BGP). Cisco Systems and T. J. Watson Research Center, IBM Corp., June 1989. RFC-II05.]] Google ScholarDigital Library
- 34.MCCAN~'E, S. Receiver-driven Layered Multicast. PhD thesis, University of California, Berkeley, Dec. 1996.]]Google Scholar
- 35.MCCANNF., S., AND JACOBSON, V. vic: A Flexible Framework for Packet Video. In Proceedings of A CM Multimedia '95 (Nov. 1995), ACM.]] Google ScholarDigital Library
- 36.McQumr. AN, J., ET AL. A New Routing Algorithm for the ARPANET. IEEE 7h~nscations on Networking (May 1980).]]Google Scholar
- 37.MOCKAPETRIS, P. Domain Names - Implementation and Specification. SRI International, Menlo Park, CA, Nov. 1987. RFC-1035.]] Google ScholarDigital Library
- 38.PAXSON, V. End-to-End Routing Behavior in the Internet. In Proc. A CM SIGCOMM '96 (Aug. 1996).]] Google ScholarDigital Library
- 39.POINTCAST INC. PointCast Home Page. http://www.p oint cast. com.]]Google Scholar
- 40.RAMAN, S., AND MCCANr~E, S. Scalable Data Naming for Application Level Framing in Reliable Multicast. In Proceedings of ACM Multimedia '98 (Bristol, UK, Sept. 1998), ACM.]] Google ScholarDigital Library
- 41.REALNETWORKS, INC. RealPlayer. http://www.real.com/.]]Google Scholar
- 42.REJAm, R., HANDLZY, M., AND ESTR~N, D. RAP: An Endto-end Rate-based Congestion Control Mechanism for Realtime Streams in the lnternet. IEEE lnfocom (1999).]]Google Scholar
- 43.PdVEST, R. The MD5 Message-Digest Algorithm. MIT Laboratory for Computer Science and RSA Data Security, Inc., 1992. RFC-1321.]]Google Scholar
- 44.SCHOONER, E. M. A Multicast User Directory Service for Synchronous Rendezvous. Tech. rep., California Institute of Technology, Pasadena, CA, August 1996.]] Google ScholarDigital Library
- 45.SCHULZRINNE, H., CASNE,~, S., FREDERICK, R., AND JACOB- SON, V. RTP: A Transport Protocol for Real-Time Applications. Internet Engineering Task Force, Audio-Video Transport Working Group, Jan. 1996. I~FC-1889.]]Google Scholar
- 46.SHARMA, P., ESTRIN, D., FLOYD, S., AND JACOBSON, V. Scalable Timers for Soft State Protocols. In Proceedings IEEE ~nfocom '97 (Kobe, Japan, 1997).]] Google ScholarDigital Library
- 47.STOICA, I., ZHANC, H., AND NO, T. S. E. A Hierarchical Fair Service Curve Algorithm for Link-Sharing, Real-Time and Priority Service. In Proceedings of SIGCOMM 1997 (Cannes, France, Sep 1997), ACM.]] Google ScholarDigital Library
- 48.TURLETTI, T., AND BOLOT, J.-C. Issues with Multicast Video Distribution in Heterogeneous Packet Networks. In Proceedings of the Sixth International Workshop on Packet Video (Portland, OR., Sept. 1994).]]Google Scholar
- 49.Digital Subscriber Signalling System No. 1 (DSS I) - ISDN User-Network Interface Layer 3 Specification for Basic Call Control, 1993. ITU-T Recommendation Q.931.]]Google Scholar
- 50.Procedures for Establishing Communication between Three or more Audiovisual Terminals using Digital Channels up to 1920 kbit/s, 1997. ITU-T Recommendation H.243.]]Google Scholar
- 51.WALDSPUR(}ER, C. A., AND WEIHL, W. E. Lottery Scheduling: Flexible Proportional-Share Resource Management. In First Symposium on Operating Systems Design and Implementation (OSDI) (1995), USENIX Association, pp. 1-11.]] Google ScholarDigital Library
- 52.WALOSPUROER, C. A., AND WEmL, W. E. Stride Scheduling: Deterministic Proportional-Share Resource. Tech. Rep. MIT/LCS/TM-528, MIT Laboratory for Computer Science, Cambridge, MA, June 1995.]] Google ScholarDigital Library
- 53.ZHANG, L., DEF, RIN~, S., ESTRIN, D., SHENKER, S., AND ZAPPALA, D. R. SVP: A New Resource ReSerVation Protocol. IEEE Network Magazine (Sept. 1993), 8-18.]]Google Scholar
Index Terms
- A model, analysis, and protocol framework for soft state-based communication
Recommendations
A model, analysis, and protocol framework for soft state-based communication
"Soft state" is an often cited yet vague concept in network protocol design in which two or more network entities intercommunicate in a loosely coupled, often anonymous fashion. Researchers often define this concept operationally (if at all) rather than ...
On the Robustness of Soft State Protocols
ICNP '04: Proceedings of the 12th IEEE International Conference on Network ProtocolsSoft state has been a mantra of Internet protocol design for the past decade. System designers build protocols that implement soft state mechanisms based on intuition or on qualitative arguments that the design is "better", yet there has never been a ...
Analysis of Internet multicast traffic performance considering multicast routing protocol
ICNP '00: Proceedings of the 2000 International Conference on Network ProtocolsRecently audio and video delivery services are widely spread in the Internet. In order to deliver these data to multiple receivers at the same time, the multicast technologies are indispensable. In such a situation, the performance analysis of multicast ...
Comments