Title :
Quantized EDF scheduling in a stochastic environment
Author :
J.P. Hansen; Haifeng Zhu;J. Lehoczky;R. Rajkumar
fDate :
6/24/1905 12:00:00 AM
Abstract :
The effects of quantization on EDF (Earliest Deadline first) queues with a stochastic traffic model are presented. In a quantized EDF queue, jobs have limited fidelity in representing their relative deadlines. It is shown that most of the benefits of full EDF can be achieved with relative deadlines expressed in as little as three bits. The implications of this are that QoS networking technologies such as DiffServ can be used to implement near EDF performance while still using no more than the standard set of bits in DS/TOS field.
Keywords :
"Stochastic processes","Telecommunication traffic","Traffic control","Video compression","Processor scheduling","Queueing analysis","Real time systems","Systems engineering and theory","Statistics"
Conference_Titel :
Parallel and Distributed Processing Symposium., Proceedings International, IPDPS 2002, Abstracts and CD-ROM
Print_ISBN :
0-7695-1573-8
DOI :
10.1109/IPDPS.2002.1016476