Previous |  Up |  Next


active queue management (AQM); drop function; finite buffer; queue size
In the article finite-buffer queueing systems of the $M/M/1/N$ type with queue size controlled by AQM algorithms are considered, separately for single and batch arrivals. In the latter case two different acceptance strategies: WBAS (Whole Batch Acceptance Strategy) and PBAS (Partial Batch Acceptance Strategy) are distinguished. Three essential characteristics of the system are investigated: the stationary queue-size distribution, the number of consecutively dropped packets (batches of packets) and the time between two successive accepted packets (batches of packets). For these characteristics the formulae which can be easily numerically treated are derived. Numerical results obtained for three sample dropping functions are attached as well.
[1] Athuraliya, S., Low, S. H., Li, V. H., Qinghe, Y.: REM: active queue management. IEEE Network 15 (2001), 3, 48–53. DOI 10.1109/65.923940
[2] Bonald, T., May, M., Bolot, J. Ch.: Analytic evaluation of RED performance. In: Proc. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies 3 (2000), pp. 1415–1424.
[3] Chrost, L., Brachman, A., Chydzinski, A.: On the performance of AQM algoritms with small buffers. Comput. Network CCIS 39 (2009), 168–173. DOI 10.1007/978-3-642-02671-3_20
[4] Floyd, S., Jacobson, V.: Random early detection gateways for congestion avoidance. IEEE ACM T. Network 1 (1993), 4, 397–412. DOI 10.1109/90.251892
[5] Floyd, S.: Recommendations on using the gentle variant of RED., March 2000.
[6] Frey, A., Takahashi, Y.: An $M^{X}/GI/1/N$ queue with close-down and vacation times. J. Appl. Math. Stoch. Anal. 12 (1999), 1, 63–83. DOI 10.1155/S1048953399000064 | MR 1690183
[7] Hao, W., Wei, Y.: An extended $GI^{X}/M/1/N$ queueing model for evaluating the performance of AQM algorithms with aggregate traffic. Lect. Notes Comput. Sci. 3619 (2005), 395–404. DOI 10.1007/11534310_43
[8] Rosolen, V., Bonaventure, O., Leduc, G.: A RED discard strategy for ATM networks and its performance evaluation with TCP/IP traffic. Comput. Comm. R. 29 (1999), 3, 23–43. DOI 10.1145/505724.505728
[9] Sun, L., Wang, L.: A novel RED scheme with preferential dynamic threshold deployment. In: Computational Intelligence and Security Workshops 2007, pp. 854–857.
[10] Suresh, S., Gol, O.: Congestion management of self similar IP traffic – application of the RED scheme. In: Wireless and Optical Communications Networks, Second IFIP International Conference 2005, pp. 372–376.
[11] Takagi, H.: Queueing analysis, Volume 1: Vacation and priority systems, Part 1. North–Holland, Amsterdam – London – New York – Tokyo 1983. MR 1149382
[12] Xiong, N., Yang, Y., Defago, X., He, Y.: LRC-RED: A self-tuning robust and adaptive AQM scheme. In: Sixth International Conference on Parallel and Distributed Computing Applications and Technologies 2005, pp. 655–659.
[13] Zhou, K., Yeung, K. L., Li, V. O. K.: Nonlinear RED: A simple yet efficient active queue management scheme. Comput. Network 50 (2006), 18, 3784–3794. DOI 10.1016/j.comnet.2006.04.007 | Zbl 1103.68364
Partner of
EuDML logo