ABSTRACT
In recent years, a number of link scheduling algorithms have been proposed that greatly improve upon traditional FIFO scheduling in being able to assure rate and delay bounds for individual sessions. However, they cannot be easily deployed in a backbone environment with thousands of sessions, as their complexity increases with the number of sessions. In this paper, we propose and analyze an approach that uses a simple buffer management scheme to provide rate guarantees to individual flows (or to a set of flows) multiplexed into a common FIFO queue. We establish the buffer allocation requirements to achieve these rate guarantees and study the trade-off between the achievable link utilization and the buffer size required with the proposed scheme. The aspect of fair access to excess bandwidth is also addressed, and its mapping onto a buffer allocation rule is investigated. Numerical examples are provided that illustrate the performance of the proposed schemes. Finally, a scalable architecture for QoS provisioning is presented that integrates the proposed buffer management scheme with WFQ scheduling that uses a small number of queues.
- 1.A. K. Choudhury and E. L. Hahne. Dynamic queue length thrcsholds in a shared memory ATM switch. In Proceedings of INFOCOM, pages 679-687, San Francisco, CA, April 1996. Google ScholarDigital Library
- 2.I. Cidon, R. Gu&in, and A. Khamisy. Protective buffer management policies. IEEE/ACM Trans. Networking, 2(3):240- 246, June 1994. Google ScholarDigital Library
- 3.S. Floyd and V. Jacobson. Random early detection gateways for congestion avoidance. IEEE/ACM Trans. Networ/,qng, 1(4):397-413, August 1993. Google ScholarDigital Library
- 4.L. Georgiadis, R. Gu&in, V. Peris, and K. N. Sivarajan. Efficient network QoS provisioning based on per node traffic shaping. IEEE/ACM Trans. Networking, 4(4):482-501, August 1996. Google ScholarDigital Library
- 5.D. Lin and R. Morris. Dynamics of random early detection. In Proceedings of SIGCOMM, pages 127-137, Sophia Antipolis, France, September 1997. Google ScholarDigital Library
- 6.A. K. J. Parekh. A Generalized Processor Sharing Approach to Flow Control in Integrated Services Networks. PhD thesis, Laboratory for Information and Decision Systems, Massachusetts Institute of Technology, Cambridge, MA 02139, February 1992. No. LIDS-TH-2089.Google Scholar
- 7.A. Romanow and S. Floyd. Dynamics of TCP traffic over ATM networks. IEEE dl Sel. Areas Commun., 13(4):633-641, May 1995. Google ScholarDigital Library
- 8.S. Suri, G. Varghese, and G. Chandranmenon. Leap forward virtual clock: A new fair queueing scheme with guaranteed dealy and throughput fairness. In Proceedings oflNFOCOM, pages 558-566, Kobe, Japan, April 1997. Google ScholarDigital Library
- 9.J. Turner. Maintaining high throughput during overload in ATM switches. In Proceedings of lNFOCOM, pages 287- 295, San Francisco, CA, April 1996. Google ScholarDigital Library
- 10.D. Wrege and J. Liebeherr. A near-optimal packet scheduler fbr QoS networks. In Proceedings oflNFOCOM, pages 577- 585, Kobe, Japan, April 1997. Google ScholarDigital Library
Index Terms
- Scalable QoS provision through buffer management
Recommendations
Scalable QoS provision through buffer management
In recent years, a number of link scheduling algorithms have been proposed that greatly improve upon traditional FIFO scheduling in being able to assure rate and delay bounds for individual sessions. However, they cannot be easily deployed in a backbone ...
Performance evaluation of prioritized scheduling with buffer management for differentiated services architectures
Differentiated services (DiffServ) is an architecture for the Internet in which various applications are supported using a simple classification scheme. Packets entering the DiffServ domain are marked depending on the packets' class. In this paper we ...
Rate allocation and buffer management for differentiated services
Special issue: Towards a new internet architectureA novel algorithm for buffer management and rate allocation is presented for providing loss and delay differentiation for traffic classes at a network router. The algorithm, called JoBS, provides delay and loss differentiation independently at each node,...
Comments