Consider the processor sharing node (PSN) of Figure 3.1 which is, for our purposes, a switch output port as well. A PSN consists of a bank of synchronized, discrete-time FIFO queues sharing an output link with a service capacity of one cell per unit time (c cells/s). Let N be the number of FIFO queues that have bandwidth guarantees, e.g., FIFO queues handling CBR or VBR traffic.
KeywordsBandwidth Allotment FIFO Queue Service Bandwidth Generalize Processor Sharing Virtual Time
Unable to display preview. Download preview PDF.
Discussion and References
- S.J. Golestani. A self-clocked fair queueing scheme for broadband applications. In IEEE INFOCOM Proc., pages 636–646, 1994.Google Scholar
- H. Zhang and S. Keshay. Comparison of rate-based service disciplines. In Proc. ACM SIGCOMM, pages 2–12, 1993.Google Scholar
- C.R. Kalmanek, H. Kanakia, and S. Keshay. Rate controlled servers for very high-speed networks. In Proc. IEEE Globecom, 1990.Google Scholar
- R.L. Cruz. Service burstiness and dynamic burstiness measures: a framework. J. High-Speed Networks, Vol. 1, No. 2:pages 105–127, 1992.Google Scholar
- H. Ohsaki, M. Murata, H. Suzuki, C. Ikeda, and H. Miyahara. Rate-based congestion control for ATM networks. ACM SIGCOMM, Computer Communication Review,pages pages 60–72, 1995.Google Scholar
- R. Cruz. Quality of service guarantees in virtual circuit switched networks. IEEE JSAC, Vol. 13, No. 6:pages 10481056, Aug. 1995.Google Scholar
- H. Zhang. Service disciplines for guaranteed performance service in packet-switching networks. Proc. of the IEEE, Vol. 83, No. 10, Oct. 1995.Google Scholar