Scheduling Algorithms for RealTime Multiprocessor Systems NICK MATHEW
Outline Multiprocessor Real-Time Systems Scheduling Algorithms Multiprocessor Scheduling P-Fair Scheduling Algorithm
Multiprocessor Real-Time Systems High computational demands Telemedicine, Tracking Systems Traditional Scheduling Assumptions Single Processor
Scheduling Algorithms Fixed-Priority Algorithm Rate Monotonic Scheduling (RMS) Optimal fixed priority algorithm At least 0.693 system utilization
Dynamic Priority Algorithm Earliest Deadline First (EDF)
Optimal dynamic-priority algorithm
* Liu & Layland, 1978
Multiprocessor Scheduling "Schedule ji jobs with length li over m processors such that none overlap“
Traditional Scheduling Algorithms EDF, RMS – low processor utilization
Multiprocessor Scheduling Approaches Partitioning Uniprocessor Low processor utilization Bin Packing NP-hard problem
Global Scheduling
Single priority-ordered queue Low processor utlization
* J. Carpenter, S. Funk, et al [2]
P-Fair Scheduling Algorithm Proportionate-Fair Scheduling Global Scheduling Algorithm Each task executes at uniform rate
* Baruah et al.
P-Fair Scheduling Algorithm Each task with weight w would receive w.L units of
processor time over interval L
Each task allocated resource for x.e time units in
each interval
Advantages Avoids bin-packing problem Handle dynamic events task leaving & joining a system
Optimal
* J. Carpenter, S. Funk, et al [2]
Disadvantages Degraded processor affinity Preemption, Migration limit effectiveness
* J. Carpenter, S. Funk, et al [2]
Conclusion Only known optimal approach (Srinivasan, 2003) Basis for PD & PF algorithms Other scheduling algorithms Genetic Algorithms Approximation Algorithms Memetic Algorithms
References [1] K. Ramamritham, J.A. Stankovic, and P.-F. Shiah, "Efficient Scheduling Algorithms for Real-Time Multiprocessor Systems," IEEE Trans. Parallel and Distributed Systems, vol. 1, no. 2, pp. 184194, Apr. 1990. [2] J. Carpenter, S. Funk, et al. A categorization of real-time multiprocessor scheduling problems and algorithms. In J. Y. Leung, editor, Handbook on Scheduling Algorithms, Methods, and Models, page 30.130.19 , 2004. [3] S. Baruah, N. Cohen, C. Plaxton and D. Varvel, "Proportionate Progress: A Notion of Fairness in Resource Allocation", Algorithmica 15(6), pgs. 600-625, 1996.
Questions ?