2015 IEEE International Conference on Data Science and Data Intensive Systems (DSDIS) (2015)
Dec. 11, 2015 to Dec. 13, 2015
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/DSDIS.2015.66
Supercomputer's fast processing speed provides a great convenience to the scientists who dealing with extremely large data sets. The next generation of "exascale" supercomputers could provide accurate simulation results in the area of automobile industry, aerospace and even nuclear fusion reactors for the very first time. However, the energy cost of super-computing is "super" expensive with a total electricity bill of 9 million dollars per year. Thus, Conserving energy or increase the energy efficiency are becoming more critical. Many researchers are looking into this problem and try to conserve energy by incorporating DVFS technique into their specific methods. However, this approach is limited especially when the workload is high. In this paper, we developed a power-aware job scheduler by applying rule based control method as well as real power and speedup profiles to improve power efficiency while maintain the power constraints. The intensive simulation results shown that our proposed method is able to achieve the maximum utilization of computing resources, in the meantime, keep the energy cost under the threshold. Moreover, by introducing a Power Performance Factor (PPF) based on the real power and speedup profiles, we are able to increase the power efficiency up to 75%.
Power demand, Supercomputers, Mathematical model, Scheduling algorithms, Job shop scheduling
R. Wang, D. Tiwari and J. Wang, "Low Power Job Scheduler for Supercomputers: A Rule-Based Power-Aware Scheduler," 2015 IEEE International Conference on Data Science and Data Intensive Systems (DSDIS), Sydney, Australia, 2015, pp. 732-733.