Energy Efficient Scheduling of Parallelizable Jobs

In this paper, we consider scheduling parallelizable jobs in the non-clairvoyant speed scaling setting to minimize the objective of weighted flow time plus energy. Previously, strong lower bounds were shown on this model in the unweighted setting even when the algorithm is given a constant amount of resource augmentation over the optimal solution. However, these lower bounds were given only for certain families of algorithms that do not recognize the parallelizability of alive jobs. In this work, we circumvent previous lower bounds shown and give a scalable algorithm under the natural assumption that the algorithm can know the current parallelizability of a job. When a general power function is considered, this is also the first algorithm that has a constant competitive ratio for the problem using any amount of resource augmentation.

[1]  Kirk Pruhs,et al.  Scalably scheduling processes with arbitrary speedup curves , 2009, TALG.

[2]  Prudence W. H. Wong,et al.  Energy efficient online deadline scheduling , 2007, SODA '07.

[3]  Lap-Kei Lee,et al.  Non-clairvoyant Speed Scaling for Weighted Flow Time , 2010, ESA.

[4]  Kirk Pruhs,et al.  Scheduling jobs with varying parallelizability to reduce variance , 2010, SPAA '10.

[5]  K. Nguyen.  Lagrangian Duality in Online Scheduling with Resource Augmentation and Speed Scaling , 2013, ESA.

[6]  Nikhil Bansal,et al.  Weighted flow time does not admit O(1)-competitive algorithms , 2009, SODA.

[7]  Kirk Pruhs,et al.  Online Primal-Dual for Non-linear Optimization with Applications to Speed Scaling , 2011, WAOA.

[8]  Nguyen Kim Thang.  Lagrangian Duality based Algorithms in Online Energy-Efficient Scheduling , 2016, SWAT.

[9]  Trevor N. Mudge,et al.  Power: A First-Class Architectural Design Constraint , 2001, Computer.

[10]  Kyle Fox,et al.  Energy Efficient Scheduling of Parallelizable Jobs , 2013, SODA.

[11]  Lachlan L. H. Andrew,et al.  Optimal speed scaling under arbitrary power functions , 2009, SIGMETRICS Perform. Evaluation Rev..

[12]  Nikhil R. Devanur,et al.  Primal Dual Gives Almost Optimal Energy-Efficient Online Algorithms , 2014, ACM Trans. Algorithms.

[13]  Kirk Pruhs,et al.  A tutorial on amortized local competitiveness in online scheduling , 2011, SIGA.

[14]  Nathan R. Tallent,et al.  Effective performance measurement and analysis of multithreaded applications , 2009, PPoPP '09.

[15]  Jesús Labarta,et al.  Improving processor allocation through run-time measured efficiency , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[16]  Manish Gupta,et al.  Power-Aware Microarchitecture: Design and Modeling Challenges for Next-Generation Microprocessors , 2000, IEEE Micro.

[17]  Kirk Pruhs,et al.  Average Rate Speed Scaling , 2008, Algorithmica.

[18]  Lap-Kei Lee,et al.  Scheduling for Weighted Flow Time and Energy with Rejection Penalty , 2011, STACS.

[19]  Kirk Pruhs,et al.  Scalably Scheduling Power-Heterogeneous Processors , 2010, ICALP.

[20]  Nikhil Bansal,et al.  Scheduling for Speed Bounded Processors , 2008, ICALP.

[21]  Kirk Pruhs,et al.  Nonclairvoyant Speed Scaling for Flow and Energy , 2010, Algorithmica.

[22]  Alan H. Karp,et al.  Measuring parallel processor performance , 1990, CACM.

[23]  Susanne Albers,et al.  Energy-efficient algorithms for flow time minimization , 2006, STACS.

[24]  Tommi Mikkonen,et al.  Flexibility as a Design Driver , 2001, Computer.

[25]  Kirk Pruhs,et al.  Speed scaling for weighted flow time , 2007, SODA '07.

[26]  Prudence W. H. Wong,et al.  Sleep with Guilt and Work Faster to Minimize Flow Plus Energy , 2009, ICALP.

[27]  Stefano Leonardi,et al.  Approximating total flow time on parallel machines , 1997, STOC '97.

[28]  Prudence W. H. Wong,et al.  Optimizing throughput and energy in online deadline scheduling , 2009, TALG.

[29]  P. Sadayappan,et al.  Moldable Parallel Job Scheduling Using Job Efficiency: An Iterative Approach , 2006, JSSPP.

[30]  Bala Kalyanasundaram,et al.  Speed is as powerful as clairvoyance , 2000, JACM.

[31]  Kirk Pruhs,et al.  Speed scaling to manage energy and temperature , 2007, JACM.

[32]  Tak Wah Lam,et al.  Tradeoff between Energy and Throughput for Online Deadline Scheduling , 2010, WAOA.

[33]  Prudence W. H. Wong,et al.  Deadline scheduling and power management for speed bounded processors , 2010, Theor. Comput. Sci..

[34]  Prudence W. H. Wong,et al.  Competitive non-migratory scheduling for flow time and energy , 2008, SPAA '08.

[35]  Kirk Pruhs,et al.  SelfishMigrate: A Scalable Algorithm for Non-clairvoyantly Scheduling Heterogeneous Processors , 2014, 2014 IEEE 55th Annual Symposium on Foundations of Computer Science.

[36]  Kirk Pruhs,et al.  Speed scaling of processes with arbitrary speedup curves on a multiprocessor , 2009, SPAA '09.

[37]  Kirk Pruhs,et al.  Scheduling heterogeneous processors isn't as easy as you think , 2012, SODA.

[38]  N. Bansal,et al.  Speed scaling with an arbitrary power function , 2009, SODA 2009.

[39]  Benjamin Moseley,et al.  Online scalable scheduling for the lk-norms of flow time without conservation of work , 2011, SODA '11.

[40]  P. Sadayappan,et al.  Effective Selection of Partition Sizes for Moldable Scheduling of Parallel Jobs , 2002, HiPC.

[41]  Yuxiong He,et al.  Speed Scaling for Energy and Performance with Instantaneous Parallelism , 2011, TAPAS.

[42]  Jeff Edmonds,et al.  Scheduling in the dark , 1999, STOC '99.

[43]  P. Sadayappan,et al.  A Robust Scheduling Strategy for Moldable Scheduling of Parallel Jobs. , 2003 .

[44]  Nikhil Bansal,et al.  Better Scalable Algorithms for Broadcast Scheduling , 2014, TALG.

[45]  Kirk Pruhs,et al.  Improved Bounds for Speed Scaling in Devices Obeying the Cube-Root Rule , 2009, ICALP.

[46]  Prudence W. H. Wong,et al.  Energy Efficient Deadline Scheduling in Two Processor Systems , 2007, ISAAC.

[47]  Yuxiong He,et al.  Adaptive work stealing with parallelism feedback , 2007, PPoPP.

[48]  Thu D. Nguyen,et al.  Using Runtime Measured Workload Characteristics in Parallel Processor Scheduling , 1996, JSSPP.

[49]  Prudence W. H. Wong,et al.  Improved multi-processor scheduling for flow time and energy , 2012, J. Sched..

[50]  Xiaotie Deng,et al.  Non-Clairvoyant Multiprocessor Scheduling of Jobs with Changing Execution Characteristics , 2003, J. Sched..

[51]  Amit Kumar,et al.  Resource augmentation for weighted flow-time explained by dual fitting , 2012, SODA.

[52]  Nguyen Kim Thang,et al.  Lagrangian Duality in Online Scheduling with Resource Augmentation and Speed Scaling , 2013, ESA.