%0 Conference Paper %1 Tian:2009:SOC:1531743.1531752 %A Tian, Kai %A Jiang, Yunlian %A Shen, Xipeng %B Proceedings of the 6th ACM conference on Computing frontiers %C New York, NY, USA %D 2009 %I ACM %K cmp multiprocessor scheduling %P 41--50 %R 10.1145/1531743.1531752 %T A study on optimally co-scheduling jobs of different lengths on chip multiprocessors %X Cache sharing in Chip Multiprocessors brings cache contention among corunning processes, which often causes considerable degradation of program performance and system fairness. Recent studies have seen the effectiveness of job co-scheduling in alleviating the contention. But finding optimal schedules is challenging. Previous explorations tackle the problem under highly constrained settings. In this work, we show that relaxing those constraints, particularly the assumptions on job lengths and reschedulings, increases the complexity of the problem significantly. Subsequently, we propose a series of algorithms to compute or approximate the optimal schedules in the more general setting.</p> <p>Specifically, we propose an A*-based approach to accelerating the search for optimal schedules by as much as several orders of magnitude. For large problems, we design and evaluate two approximation algorithms, A*-cluster and local-matching algorithms, to effectively approximate the optimal schedules with good accuracy and high scalability. This study contributes better understanding to the optimal co-scheduling problem, facilitates the evaluation of co-scheduling systems, and offers insights and techniques for the development of practical co-scheduling algorithms. %@ 978-1-60558-413-3

@inproceedings{Tian:2009:SOC:1531743.1531752, abstract = {Cache sharing in Chip Multiprocessors brings cache contention among corunning processes, which often causes considerable degradation of program performance and system fairness. Recent studies have seen the effectiveness of job co-scheduling in alleviating the contention. But finding optimal schedules is challenging. Previous explorations tackle the problem under highly constrained settings. In this work, we show that relaxing those constraints, particularly the assumptions on job lengths and reschedulings, increases the complexity of the problem significantly. Subsequently, we propose a series of algorithms to compute or approximate the optimal schedules in the more general setting.</p> <p>Specifically, we propose an A*-based approach to accelerating the search for optimal schedules by as much as several orders of magnitude. For large problems, we design and evaluate two approximation algorithms, A*-cluster and local-matching algorithms, to effectively approximate the optimal schedules with good accuracy and high scalability. This study contributes better understanding to the optimal co-scheduling problem, facilitates the evaluation of co-scheduling systems, and offers insights and techniques for the development of practical co-scheduling algorithms.}, acmid = {1531752}, added-at = {2012-09-21T22:23:22.000+0200}, address = {New York, NY, USA}, author = {Tian, Kai and Jiang, Yunlian and Shen, Xipeng}, biburl = {https://www.bibsonomy.org/bibtex/2fb1e9754bf2176d9576e1b704f1e3863/ytyoun}, booktitle = {Proceedings of the 6th ACM conference on Computing frontiers}, doi = {10.1145/1531743.1531752}, interhash = {a42f415acd71750eea504b0e57c33c0d}, intrahash = {fb1e9754bf2176d9576e1b704f1e3863}, isbn = {978-1-60558-413-3}, keywords = {cmp multiprocessor scheduling}, location = {Ischia, Italy}, numpages = {10}, pages = {41--50}, publisher = {ACM}, series = {CF '09}, timestamp = {2012-09-21T22:23:22.000+0200}, title = {A study on optimally co-scheduling jobs of different lengths on chip multiprocessors}, year = 2009 }