Title :
Energy modeling of supercomputers and large-scale scientific applications
Author :
Pakin, Scott ; Lang, Michael
Author_Institution :
Comput., Comput., & Stat. Sci. Div., Los Alamos Nat. Lab., Los Alamos, NM, USA
Abstract :
We derive models of time, power, and energy as a function of frequency and demonstrate that these models accurately reflect measurements of large, parallel, scientific applications. We then apply these models to explain how sensitive the various applications are to CPU frequency; to analyze the compute-boundedness (frequency sensitivity) of complete, production workloads running on three of the world´s fastest supercomputers; to predict how changes in system power properties will impact applications; and to determine what CPU frequencies offer the best energy, EDP, and ED2P savings.
Keywords :
parallel machines; power aware computing; CPU frequency; ED2P savings; EDP; compute-boundedness; energy modeling; frequency sensitivity; large-scale scientific application; supercomputers; Computational modeling; Energy consumption; Equations; Frequency measurement; Mathematical model; Power measurement; Supercomputers;
Conference_Titel :
Green Computing Conference (IGCC), 2013 International
Conference_Location :
Arlington, VA
DOI :
10.1109/IGCC.2013.6604494
