Title :
Modeling Power and Energy Usage of HPC Kernels
Author :
Tiwari, Ananta ; Laurenzano, Michael A. ; Carrington, Laura ; Snavely, Allan
Author_Institution :
Performance Modeling & Characterization Lab., San Diego Supercomput. Center, San Diego, CA, USA
Abstract :
Compute intensive kernels make up the majority of execution time in HPC applications. Therefore, many of the power draw and energy consumption traits of HPC applications can be characterized in terms of the power draw and energy consumption of these constituent kernels. Given that power and energy-related constraints have emerged as major design impediments for exascale systems, it is crucial to develop a greater understanding of how kernels behave in terms of power/energy when subjected to different compiler-based optimizations and different hardware settings. In this work, we develop CPU and DIMM power and energy models for three extensively utilized HPC kernels by training artificial neural networks. These networks are trained using empirical data gathered on the target architecture. The models utilize kernel-specific compiler-based optimization parameters and hard-ware tunables as inputs and make predictions for the power draw rate and energy consumption of system components. The resulting power draw and energy usage predictions have an absolute error rate that averages less than 5.5% for three important kernels - matrix multiplication (MM), stencil computation and LU factorization.
Keywords :
neural nets; operating system kernels; power aware computing; CPU; DIMM; HPC kernels; LU factorization; artificial neural networks; compiler-based optimizations; compute intensive kernels; energy consumption; energy usage; exascale systems; hardware tunables; kernel-specific compiler-based optimization parameters; matrix multiplication; power usage; stencil computation; Computational modeling; Kernel; Mathematical model; Optimization; Power measurement; Predictive models; Training;
Conference_Titel :
Parallel and Distributed Processing Symposium Workshops & PhD Forum (IPDPSW), 2012 IEEE 26th International
Conference_Location :
Shanghai
Print_ISBN :
978-1-4673-0974-5
DOI :
10.1109/IPDPSW.2012.121