Lixia Zhang
Abstract
One of the challenging research issues in building high-speed packet-switched networks is how to control the transmission rate of statistical data flows. This paper describes a new traffic control algorithm, VirtualClock, for high-speed network applications. VirtualClock monitors the average transmission rate of statistical data flows and provides every flow with guaranteed throughput and low queueing delay. It provides firewall protection among individual flows, as in a TDM system, while retaining the statistical multiplexing advantages of packet switching. Simulation results show that the VirtualClock algorithm meets all its design goals.
- 1 CHERITON, D. VMTP: A transport protocol for the next generation of communication systems. In the Proceedings ofSIGCOMM'86 (Stowe, Vt., Aug. 1986). Google Scholar
- 2 CLARK, D., LAMBERT, M., AND ZHANG, L. A high throughput bulk data transfer protocol. In the Proceedings of SIGCOMM'$7 (Stowe, Vt., Aug. 1987).Google Scholar
- 3 DEMERS, A., KESHAV, S., AND SHENKER, S. Analysis and simulation of a fair queueing algorithm. In the Proceedings of SIGCOMM'89 (Austin, Tx., Sept. 1989). Google Scholar
- 4 DESCLOUS, A. Contention probabilities in packet switching networks with strung input processes. In Proceedings of the 12th Teletraffic Congress (Turin, 1988), 51A. 1.1-1.7.Google Scholar
- 5 FERRARI, D. Real-time communication in packet-switching wide-area networks. Tech. Rep. TR~89-022, International Computer Science Institute, Berkeley, Calif., May 1989.Google Scholar
- 6 FERRARI, D., AND VERMA D. A scheme ibr real-time channel establishment in wide-area networks. IEEE J. Selected Areas Commun. 8, 3 (Apr. 1990), 368-389.Google Scholar
- 7 FERRARI, D. Client requirements for real-time communication services. IEEE Commun. Mag. 28, 11 (Nov. 1990), 65-72.Google Scholar
- 8 HAHNE, E. L. Round robin scheduling for fair flow control in data communication networks. Ph.D. thesis, MIT, Dec. 1986.Google Scholar
- 9 GROVER, G., AND BYIARATH-KUMAR, K. Windows in the sky--Flow control in SNA networks with satellite links. IBM Syst. J. 22, 4 (1983), 451-463.Google Scholar
- 10 JAIN, e., AND ROUTHIER, S. Packet trains--Measurements and a new model for computer network traffic. IEEE J. Selected Areas Commun. SAC4, 6 (Sept. 1986), 986-995.Google Scholar
- 11 JAIN, e., RAMAKRISHNAN, K., AND CHIU, D. Congestion avoidance in computer networks with a connectionless network layer. In Ir~r~ovation6 in lnternetworking, Artech House, 1988. Google Scholar
- 12 KALMANEK, C., KANAKIA, H., AND KESHAV, S. Rate controlled servers for very high speed networks. In the Proceedings of GLOBCOM'90 (San Diego, Cali~, Dec. 1990), 12-20.Google Scholar
- 13 LAMBERT, M. An end-point adaptive rate control strategy for the NETBLT protocol. Unpublished research notes, MIT Lab for Computer Science, May 1988.Google Scholar
- 14 MUKHERJI, U. A schedule-based approach for flow-control in data communication networks. Ph.D. thesis, MIT, Feb. 1986.Google Scholar
- 15 NAGLE, J. Congestion control in TCP/IP internetworks. ACM Cornput. Commun. Rev 14, 4 (Oct 1984). Google Scholar
- 16 PETR, D., DASmvA, L., AND FROST, V. Priority discarding of speech in integrated packet networks. IEEE J. Selected Areas Commun. 7, 5 (June 1989), 644-656.Google Scholar
- 17 RATHOEB, E. P Comparison of policing mechanisms for ATM networks. In Proceedings of the 3rd RACE Workshop (Paris, Oct. 1989). Paper 21.1Google Scholar
- 18 SEN, P., MAGLARIS, B., RIKLI, N., AND ANASTASSIOU, D Models for packet switching of variable-bit-rate video sources IEEE J. Selected Areas Commun. 7, 5 (June, 1989), 865-869.Google Scholar
- 19 TCP-IP mailing list. TCP-IP mailing list is a special-interest-group mailing list moderated by the Network Information Center (NIC) located at SRI. In TCP-IP mail discussion, there have been numerous observations of malfunctioning hosts m the ARPA Internet.Google Scholar
- 20 TURNER, J. New directions in communications (or Which way to the information age?). IEEE Commun. Mag. 24, 10 (Oct. 1986), 8 15.Google Scholar
- 21 TURNER, J. Personal communication, 1989.Google Scholar
- 22 TYMES, L. Routing and flow control in Tymnet. IEEE Trans. Commun. COM-29, 4 (April 1981), 392-398.Google Scholar
- 23 WEINRm, A VirtualClock and Leaky-Bucket: Flow control protocols for high-speed networks. In the Proceedings of International Workshop on Protocols for High-Speed Networks (Nov 1990).Google Scholar
- 24 ZHANG, L. Some thoughts on the packet network architecture. Comput. Commun. Rev. 17, 1&2 (Jam/Apr. 1987), 3-17 Google Scholar
- 25 ZHA~G, L. A new architecture for packet switching network protocols. Ph.D. thesis, Dept. of Electrical Engineering and Computer Science, MIT, July 1989.Google Scholar
Index Terms
- VirtualClock: a new traffic control algorithm for packet-switched networks
Recommendations
A framework for guaranteeing statistical QoS
Continuous-media traffic (i.e., audio and video) can tolerate some loss but has rigid delay constraints. A natural QoS requirement for a continuous-media connection is a prescribed limit on the fraction of traffic that exceeds an end-to-end delay ...
The token-bank leaky bucket mechanism for group connections in ATM networks
ICNP '96: Proceedings of the 1996 International Conference on Network Protocols (ICNP '96)A well accepted policing mechanism for ATM networks is the leaky bucket mechanism. The original leaky bucket has been shown to be effective for traffic with a constant bit rate, but reacts poorly to bursty traffic. Although there are some other policing ...
A framework for statistical multiplexing onto a variable-bit rate output channel
INFOCOM '95: Proceedings of the Fourteenth Annual Joint Conference of the IEEE Computer and Communication Societies (Vol. 2)-Volume - Volume 2This paper investigates soft multiplexing, an issue that arises in VP-based ATM networks. The main distinguishing feature of this form of multiplexing (as opposed to traditional statistical multiplexing) is that the output channel is a variable-bit rate ...
Reviews
Robert Bruce McLaughlin
The author puts forward a scheme to use virtual clocks to control the flow of data packets in a high-speed packet switched network. The scheme requires each user to request an allocation for the amount of traffic the user expects. By stamping the packets and then placing them in a queue based on the values of their virtual time-stamps, the author expects to get higher packet throughput. The scheme in and of itself does not provide any way to correct and detect errors. The author indicates that the scheme expects a level 4 transport layer, which means that it will not support X.25 without major modification. The scheme is similar to standard TDM schemes and to several schemes that are used and have been proposed for packet satellite transmissions. For some reason, no comparison to similar packet multiplexing is made. The paper ends with a brief review of expected performance based on limited experiments with a SUN network. The performance is improved over other schemes, but a lot of work is required before any decision on use of the method can be made.
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments