ABSTRACT
A future large-scale high-performance computing (HPC) cluster will likely be power capped since the surrounding infrastructure like power supply and cooling is constrained. For such a cluster, it may be impossible to supply thermal design power (TDP) to all components. The default power supply of current system guarantees TDP to each computing node will become unfeasible. Power capping was introduced to limit power consumption to a value below TDP, with the drawback of resulting performance limitations. We developed an alternative dynamic application-aware power scheduling (DAPS) strategy to enforce a predetermined power limit and at the same time improve the cluster-wide performance. The power scheduling decision is guided by the cap value, the hardware usage, and the application-specific performance sensitivity to power. Applying DAPS on a test platform comprising 12 computing nodes with three representative applications, we obtained a performance improvement up to 17% compared to a strategy that distributes power equally and statically across nodes.
- Axel Auweter, Arndt Bode, Matthias Brehm, Luigi Brochard, Nicolay Hammer, Herbert Huber, Raj Panda, Francois Thomas, and Torsten Wilde. 2014. A case study of energy aware scheduling on supermuc. In International Supercomputing Conference. Springer, 394--409. Google ScholarDigital Library
- Karel De Vogeleer, Gerard Memmi, Pierre Jouvelot, and Fabien Coelho. 2013. The energy/frequency convexity rule: Modeling and experimental validation on mobile devices. In International Conference on Parallel Processing and Applied Mathematics. Springer, 793--803.Google Scholar
- Qingyuan Deng, David Meisner, Abhishek Bhattacharjee, Thomas F Wenisch, and Ricardo Bianchini. 2012. Coscale: Coordinating cpu and memory system dvfs in server systems. In Proceedings of the 2012 45th Annual IEEE/ACM International Symposium on Microarchitecture. IEEE Computer Society, 143--154.Google ScholarDigital Library
- Martin Dimitrov. 2012. Intel Power Governor. https://software.intel.com/en-us/articles/intel-power-governor. (2012).Google Scholar
- Daniel A Ellsworth, Allen D Malony, Barry Rountree, and Martin Schulz. 2015. Dynamic power sharing for higher job throughput. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. ACM, 80.Google ScholarDigital Library
- Institute for Combustion Technologe. 2015. Psopen. http://www.fz-juelich.de/ias/jsc/EN/Expertise/High-Q-Club/psOpen/_node.html. (2015).Google Scholar
- SPEC HPG. 2007. SPEC MPI2007. https://www.spec.org/mpi2007/. (2007).Google Scholar
- SPEC HPG. 2012. SPEC OMP2012. https://www.spec.org/omp2012/. (2012).Google Scholar
- Tapasya Patki. 2013. librapl. https://github.com/tpatki/librapl/tree/master/libmsr. (2013).Google Scholar
- Tapasya Patki, David K Lowenthal, Barry Rountree, Martin Schulz, and Bronis R De Supinski. 2013. Exploring hardware overprovisioning in power-constrained, high performance computing. In Proceedings of the 27th international ACM conference on International conference on supercomputing. ACM, 173-182. Google ScholarDigital Library
- Tapasya Patki, David K Lowenthal, Anjana Sasidharan, Matthias Maiterth, Barry L Rountree, Martin Schulz, and Bronis R De Supinski. 2015. Practical resource management in power-constrained, high performance computing. In Proceedings of the 24th International Symposium on High-Performance Parallel and Distributed Computing. ACM, 121-132. Google ScholarDigital Library
- Barry Rountree, Dong H Ahn, Bronis R De Supinski, David K Lowenthal, and Martin Schulz. 2012. Beyond DVFS: A first look at performance under a hardware-enforced power bound. In Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), 2012 IEEE 26th International. IEEE, 947-953.Google ScholarDigital Library
- Barry Rountree, David K Lowenthal, Shelby Funk, Vincent W Freeh, Bronis R De Supinski, and Martin Schulz. 2007. Bounding energy consumption in large-scale MPI programs. In Proceedings of the 2007 ACM/IEEE conference on Supercomputing. ACM, 49. Google ScholarDigital Library
- Barry Rountree, David K Lownenthal, Bronis R De Supinski, Martin Schulz, Vincent W Freeh, and Tyler Bletsch. 2009. Adagio: making DVS practical for complex HPC applications. In Proceedings of the 23rd international conference on Supercomputing. ACM, 460-469. Google ScholarDigital Library
- Osman Sarood, Akhil Langer, Abhishek Gupta, and Laxmikant Kale. 2014. Maximizing throughput of overprovisioned hpc data centers under a strict power budget. In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis. IEEE Press, 807-818. Google ScholarDigital Library
- Robert Schöne and Daniel Hackenberg. 2011. On-line analysis of hardware performance events for workload characterization and processor frequency scaling decisions. In Proceedings of the 2nd ACM/SPEC International Conference on Performance engineering. ACM, 481-486. Google ScholarDigital Library
- Bo Wang, Dirk Schmidl, and Matthias S Müller. 2015. Evaluating the energy consumption of openmp applications on Haswell processors. In International Workshop on OpenMP. Springer, 233-246. Google ScholarCross Ref
Index Terms
Dynamic Application-aware Power Capping
Recommendations
Operation-Aware Power Capping
Euro-Par 2020: Parallel ProcessingAbstractOnce the peak power draw of a large-scale high-performance-computing (HPC) cluster exceeds the capacity of its surrounding infrastructures, the cluster’s power consumption needs to be capped to avoid hardware damage. However, power capping often ...
Power consumption evaluation of an MHD simulation with CPU power capping
CCGRID '14: Proceedings of the 14th IEEE/ACM International Symposium on Cluster, Cloud, and Grid ComputingRecently to achieve the Exa-flops next generation computer system, the power consumption becomes the important issue. On the other hand, the power consumption character of application program is not so considered now. In this study we examine the power ...
Benefits in Relaxing the Power Capping Constraint
ANDARE '17: Proceedings of the 1st Workshop on AutotuniNg and aDaptivity AppRoaches for Energy efficient HPC SystemsIn this manuscript we evaluate the impact of HW power capping mechanisms on a real scientific application composed by parallel execution. By comparing HW capping mechanism against static frequency allocation schemes we show that a speed up can be ...
Comments