Desynchronized Pfair Scheduling on Multiprocessors

Pfair scheduling, currently the only known way of optimally scheduling recurrent real-time tasks on multiprocessors, imposes certain requirements that may limit its practical implementation. In this paper, we address one such limitation — which requires processor time to always be allocated in units of fixed-sized quanta that are synchronized across processors — and determine the impact of relaxing it. We show that if this requirement is relaxed, then under an otherwise-optimal Pfair scheduling algorithm, deadlines are missed by at most one quantum only, which is sufficient to provide soft real-time guarantees. This result can be shown to extend to most prior work on Pfair scheduling: In general, tardiness bounds guaranteed by previously-proposed suboptimal Pfair algorithms are worsened by at most one quantum only. ∗ Work supported by NSF grants CCR 0204312, CCR 0309825, and CCR 0408996. 

• Type: Conference
• Year: 2005
• Where: IPPS
• Authors: UmaMaheswari C. Devi, James H. Anderson
• Keywords: Distributed And Parallel Computing, IPPS 2005, Pfair Scheduling, Pfair Scheduling Algorithm, Recurrent Real-time Tasks