Online algorithms for maximizing weighted throughput of unit jobs with temperature constraints
- 197 Downloads
We consider a temperature-aware online deadline scheduling model. The objective is to schedule a number of unit jobs, with release dates, deadlines, weights and heat contributions, to maximize the weighted throughput subject to a temperature threshold. We first give an optimally competitive randomized algorithm. Then we give a constant competitive randomized algorithm that allows a tradeoff between the maximum heat contribution of jobs and the competitiveness. Finally we consider the multiple processor case and give several tight upper and lower bounds.
KeywordsOnline algorithms Scheduling Competitive analysis Temperature Resource augmentation
We thank the anonymous reviewers for their useful comments, and Kirk Pruhs for useful discussions.
- Andelman N, Mansour Y, Zhu A (2003) Competitive queueing policies for QoS switches. In: Proceedings of 14th ACM-SIAM symposium on discrete algorithms, pp 761–770 Google Scholar
- Awerbuch B, Bartal Y, Fiat A, Rosen A (1994) Competitive non-preemptive call control. In: Proceedings of 5th ACM-SIAM symposium on discrete algorithms, pp 312–320 Google Scholar
- Coskun A, Rosing T, Whisnant K (2007) Temperature aware task scheduling in MPSoCs. In: Proc conference on design, automation and test in Europe, pp 1659–1664 Google Scholar
- Englert M, Westermann M (2007) Considering suppressed packets improves buffer management in QoS switches. In: Proceedings of 18th ACM-SIAM symposium on discrete algorithms, pp 209–218 Google Scholar
- Yang J, Zhou X, Chrobak M, Zhango Y, Jin L (2008) Dynamic thermal management through task scheduling. In: IEEE international symposium on performance analysis of systems and software, pp 191–201 Google Scholar
- Yao AC-C (1977) Probabilistic computations: Toward a unified measure of complexity. In: Proceedings of 18th IEEE symposium on foundations of computer science, pp 222–227 Google Scholar