2015 · 31 citations · 6 references
Cluster ComputingHeterogeneous ComputingEngineeringEnergy EfficiencyComputer ArchitectureParallel ComputingPresent Auto Pin+Job SchedulerHybrid Hpc WorkloadCloud SchedulingComputer EngineeringScheduling (Computing)Computer SciencePeak PerformanceCo-processorsEdge ComputingReal-time Multiprocessor SystemCloud ComputingCase StudyParallel Programming
Most applications running on supercomputers achieve only a fraction of the peak performance of the system. In this paper we analyze the performance and energy efficiency of co-scheduling one memory bandwidth bound and one compute bound application on the same node. We present auto pin+, a tool designed to monitor and optimize co-scheduling of applications. Our analysis shows that co-scheduling can improve both energy efficiency and overall throughput of a supercomputer. At best, runtime can be decreased by 28% and the energy consumption by 12%, respectively, compared to best case dedicated execution. The overall efficiency however strongly depends on the ratio of jobs available in the queue. We furthermore present a simple adaptive strategy depending on the available jobs in the queue.
6
Howard David, Eugene Gorbatov, Ulf R. Hanebutte et al. · 2010 · 568 citations
Barry Rountree, David K. Lownenthal, Bronis R. de Supinski et al. · 2009 · 282 citations
Power-aware Computing, Novel Runtime Mechanism, Engineering +12
Towards Energy Aware Scheduling for Precedence Constrained Parallel Tasks in a Cluster with DVFS
Lizhe Wang, Gregor von Laszewski, Jay Dayal et al. · 2010 · 222 citations
Massively parallel genomic sequence search on the Blue Gene/P architecture
Heshan Lin, Pavan Balaji, Ruth J. Poole et al. · 2008 · 29 citations
Cluster Computing, Blue Gene/p Architecture, Engineering +21