DocumentCode :
3601266
Title :
Cone Schedules for Processing Systems in Fluctuating Environments
Author :
Ross, Kevin ; Bambos, Nicholas ; Michailidis, George
Author_Institution :
Univ. of California Santa Cruz, Santa Cruz, CA, USA
Volume :
60
Issue :
10
fYear :
2015
Firstpage :
2710
Lastpage :
2715
Abstract :
Consider a generalized processing system with several queues, where the available service rate combinations are fluctuating over time due to reliability and availability variations, and the scheduler must respond dynamically to the workload and service availability. We establish that cone schedules, which leverage the geometry of the queueing dynamics, maximize the system throughput for a broad class of processing systems, even under adversarial arrival processes.
Keywords :
optimisation; queueing theory; telecommunication network reliability; telecommunication scheduling; adversarial arrival process; cone schedule; fluctuating environment; generalized processing system; queueing dynamics geometry; reliability variation; service availability variation; service rate; system throughput maximisation; Availability; Dynamic scheduling; Load modeling; Queueing analysis; Schedules; Stability analysis; Vectors; Availability; communication networks; processor scheduling; queueing analysis; reliability; throughput;
fLanguage :
English
Journal_Title :
Automatic Control, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9286
Type :
jour
DOI :
10.1109/TAC.2015.2398881
Filename :
7029006
Link To Document :
بازگشت