Practical trade-offs for open interconnection

ABSTRACT

There is increasing market pressure to provide support for the open interconnection of systems via general purpose protocol suites such as OSI and TCP/IP. The complexity of these protocols means that the achievement of acceptable performance is not easy. Indeed, some would claim it is impossible, and advocate lean, closed protocols. A further aspect of communications architectures in the world outside the research laboratory is that they must be well structured and modular, in order to meet the needs of orderly systems development and the provision of configurable products. This paper examines the trade-offs between these three aspects of protocol stack development: conformance to standards, reasonable performance and modularity. It finds that while a considerable amount of work has been carried out in recent times, it is apparent that we do not know yet how to achieve all three.

REFERENCES

• 1: ANDERSON, D.P. 1988. A software architecture for network communication. Proceedings of the 8th International Conference on Distributed Computer Systems (May), IEEE, New York, pp. 376-383.
• 2: Jean-Chrysostome Bolot , A. Udaya Shankar, Dynamical behavior of rate-based flow control mechanisms, ACM SIGCOMM Computer Communication Review, v.20 n.2, p.35-49, April, 1990 [doi>10.1145/378570.378633]
• 3: David A. Borman, Implementing TCP/IP on a cray computer, ACM SIGCOMM Computer Communication Review, v.19 n.2, p.11-15, April 1, 1989 [doi>10.1145/378444.378446]
• 4: BRADY, P.T. 1988. Effects on respcmse time performance using an edge-w-edge protocol in an X~.5 packet network. IEEE Network 2,4 (Jul.) pp. 45-54.
• 5: L. F. Cabrera , E. Hunter , M. J. Karels , D. A. Mosher, User-Process Communication Performance in Networks of Computers, IEEE Transactions on Software Engineering, v.14 n.1, p.38-53, January 1988 [doi>10.1109/32.4621 ]
• 6: David R. Cheriton , Willy Zwaenepoel, The distributed V kernel and its performance for diskless workstations, Proceedings of the ninth ACM symposium on Operating systems principles, p.129-140, October 10-13, 1983, Bretton Woods, New Hampshire, United States
• 7: H Y Chong, Software development and implementation of NBS class-4 transport protocol, Computer Networks and ISDN Systems, v.11 n.5, p.353-365, May 1986 [doi>10.1016/0169-7552(86)90041-3]
• 8: David D. Clark, The structuring of systems using upcalls, Proceedings of the tenth ACM symposium on Operating systems principles, p.171-180, December 1985, Orcas Island, Washington, United States
• 9: CLARK, D. D., JACOBSEN, V., ROMKEY, jo, SALWEN, H. 1989. An analysis of TCP processing overhead. IEEE Commun. 27,6 (June), pp 23-29.
• 10: D. D. Clark , M. L. Lambert , L. Zhang, NETBLT: a high throughput transport protocol, Proceedings of the ACM workshop on Frontiers in computer communications technology, p.353-359, August 11-13, 1987, Stowe, Vermont, United States
• 11: D. Clark, The design philosophy of the DARPA internet protocols, Symposium proceedings on Communications architectures and protocols, p.106-114, August 16-18, 1988, Stanford, California, United States
• 12: DEAI'ON, G. 1990. Considerations for advanced networking in the 1990's. In Proceedings of the 10th International Conference on Computer Communication, S. Ramani, H. Shrikumar and S. V. Raghavan, Eds. Narosa Publishing House, New Delhi, pp. 13-20.
• 13: FELDMEIER, D., AND BIERSACK, E. 1990. Comparison of error control protcr~ls for high bandwidth-delay prodtw.t networks. In Proceedings of the Second IFIP WG6.1/WG6.4 International Workshop on Protocols for High-Speed Networks (Palo Alto, Cal., Nov. 27-29).
• 14: FRY, M.R. 1990. The efficiency of OSI protocol stacks: A computer systems persIx~tive. In Proceedings of the lOth International Conference on Computer Communication, S. Rammfi, H. Shrikumar and S. V. Raghavan, Eds. Narosa Publishing House, New Delhi, pp. 553-557.
• 15: N. R. Howes , A. C. Weaver, Measurements of Ada Overhead in OSI-Style Communications Systems, IEEE Transactions on Software Engineering, v.15 n.12, p.1507-1517, December 1989 [doi>10.1109/32.58763 ]
• 16: N. Hutchinson , L. Peterson, Design of the x-kernel, Symposium proceedings on Communications architectures and protocols, p.65-75, August 16-18, 1988, Stanford, California, United States
• 17: L. Peterson , N. Hutchinson , S. O'Malley , M. Abbott, RPC in the x-Kernel: evaluating new design techniques, Proceedings of the twelfth ACM symposium on Operating systems principles, p.91-101, November 1989
• 18: V. Jacobson, Congestion avoidance and control, Symposium proceedings on Communications architectures and protocols, p.314-329, August 16-18, 1988, Stanford, California, United States
• 19: H. Kanakia , D. Cheriton, The VMP network adapter board (NAB): high-performance network communication for multiprocessors, Symposium proceedings on Communications architectures and protocols, p.175-187, August 16-18, 1988, Stanford, California, United States
• 20: C. Kline, Supercomputers on the internet: a case study, Proceedings of the ACM workshop on Frontiers in computer communications technology, p.27-33, August 11-13, 1987, Stowe, Vermont, United States
• 21: K. A. Lantz , William I. Nowicki , Marvin M. Theimer, An empirical study of distributed application performance, IEEE Transactions on Software Engineering, v.11 n.10, p.1162-1174, Oct. 1985 [doi>10.1109/TSE.1985.231864]
• 22: P. F. Linington, Why OSI?, Computer Networks and ISDN Systems, v.17 n.4-5, p.287-290, Oct. 10, 1989 [doi>10.1016/0169-7552(89)90041-X]
• 23: B. Liskov , D. Curtis , P. Johnson , R. Scheifer, Implementation of Argus, Proceedings of the eleventh ACM Symposium on Operating systems principles, p.111-122, November 08-11, 1987, Austin, Texas, United States
• 24: MEISTER, B. 1987. A performance study of the ISO Transport Protocol. Proceech'ngs of the 7th international Conference on Distributed Computer Systems (May), IEEE, New York, pp. 398-405.
• 25: MURRAY WOODSIE, C., AND RAMIRO MONTEALEGRE, J. 1989. TI~ effect of buffering strategies on protocol execution performance. IEEE Trans. Commun. 37,6 (June), pp. 545-554.
• 26: Bill Nowicki, Transport issues in the network file system, ACM SIGCOMM Computer Communication Review, v.19 n.2, p.16-20, April 1, 1989 [doi>10.1145/378444.378447]
• 27: O'MALLEY, S. 1990. (In discussion at Second IFIP International Workshop on Protocols for High-S~ Networks).
• 28: O'MALLEY, S. AND PETERSON, L. 1990. A highly-layered architecture for high-~ networks. In Proceedings of the Second IFIP WG6.11WG6.4 International Workshop on Protocols for High-Speed Networks (Palo Alto, Cal., Nov. 27-29).
• 29: Craig Partridge, How slow is one gigabit per second?, ACM SIGCOMM Computer Communication Review, v.20 n.1, p.44-53, Jan. 1990 [doi>10.1145/86587.86589]
• 30: PARULKAR, G.M., AND TURNER, J.S. 1990. Towards a framework for high ~ communication in a heterogenous networking environment. IEEE Network 4,2 (Mar.) pp. 19-27.
• 31: Gerald J. Popek , Bruce J. Walker, The LOCUS distributed system architecture, Massachusetts Institute of Technology, Cambridge, MA, 1986
• 32: K. K. Ramakrishnan , R. Jain, A binary feedback scheme for congestion avoidance in computer networks, ACM Transactions on Computer Systems (TOCS), v.8 n.2, p.158-181, May 1990 [doi>10.1145/78952.78955]
• 33: Robbert van Renesse , Hans van Staveren , Andrew S. Tanenbaum, Performance of the world's fastest distributed operating system, ACM SIGOPS Operating Systems Review, v.22 n.4, p.25-34, Oct. 1988 [doi>10.1145/54289.54291]
• 34: J. H. Saltzer , D. P. Reed , D. D. Clark, End-to-end arguments in system design, ACM Transactions on Computer Systems (TOCS), v.2 n.4, p.277-288, Nov. 1984 [doi>10.1145/357401.357402]
• 35: SAL'IT_,ER, J.H., CLARK, D.D., ROMKEY, J.L., AND GRAMLICH, W.C. 1985. The desktop computer as a network participant. IEEE J. Select. Areas Comm. SAC-7, 3 (May), pp. 468-478.
• 36: M. Schroeder , M. Burrows, Performance of Firefly RPC, Proceedings of the twelfth ACM symposium on Operating systems principles, p.83-90, November 1989
• 37: K. Sklower, Improving the efficiency of the OSI checksum calculation, ACM SIGCOMM Computer Communication Review, v.19 n.5, p.32-43, Oct. 1989 [doi>10.1145/74681.74684]
• 38: M. Stella Atkins, Experiments in SR with different upcall program structures, ACM Transactions on Computer Systems (TOCS), v.6 n.4, p.365-392, Nov. 1988 [doi>10.1145/48012.48041]
• 39: STERBENZ, j. P. G., AND PARULKAR, G. M. 1990. AXON: Application-orienteA lightweight transport txol~ol design. In Proceedings of the l Oth International Conference on Computer Communication, S. Ramani, H. Shrikumar and S. V. Raghavan, Eds. Narosa Publishing House, New Delhi, pp. 379-387.
• 40: STRAYER, W.T., AND WEAVER, A.C. 1988. Performance measurement of data transfer services in MAP. IEEE Network 2,3 (May) pp. 75-81.
• 41: SVOBODOVA, L. 1989A. Implementing OSI systems. IEEE J. Select. Areas Comm. SAC-7, 7 (Sept), pp. 1115-1130.
• 42: SVOBODOVA, L. 1989B. Measured performance of transport services in LAN's. Comput. Networks and ISDN Syst. 18 (Nov.), pp. 31-45.
• 43: G. Varghese , T. Lauck, Hashed and hierarchical timing wheels: data structures for the efficient implementation of a timer facility, Proceedings of the eleventh ACM Symposium on Operating systems principles, p.25-38, November 08-11, 1987, Austin, Texas, United States
• 44: Richard W. Watson , Sandy A. Mamrak, Gaining efficiency in transport services by appropriate design and implementation choices, ACM Transactions on Computer Systems (TOCS), v.5 n.2, p.97-120, May 1987 [doi>10.1145/13677.13678]
• 45: M. Young , A. Tevanian , R. Rashid , D. Golub , J. Eppinger, The duality of memory and communication in the implementation of a multiprocessor operating system, Proceedings of the eleventh ACM Symposium on Operating systems principles, p.63-76, November 08-11, 1987, Austin, Texas, United States
• 46: ZII-II~RBART, M., AND GERTEIS, W. 1990. OSI protocols for high-speed networks? In Proceedings of the lOlh International Conference on Computer Communication, S. Ramani, H. Shrikumar and S. V. Raghavan, Eds. Narosa Publishing House, New Delhi, pp. 309-314.

INDEX TERMS

Primary Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS
      C.2.0 General
          Subjects: Open Systems Interconnection reference model (OSI)

Additional Classification:
  C. Computer Systems Organization
  C.2 COMPUTER-COMMUNICATION NETWORKS

General Terms:
Design, Performance