Quantifying the Impact of GPUs on Performance and Energy Efficiency in HPC Clusters
Center for Supercomputing Applications, University of Illinois at Urbana-Champaign, Urbana, IL, USA
International Green Computing Conference, 2010, p.317-324
@conference{enos2010quantifying,
title={Quantifying the impact of GPUs on performance and energy efficiency in HPC clusters},
author={Enos, J. and Steffen, C. and Fullop, J. and Showerman, M. and Shi, G. and Esler, K. and Kindratenko, V. and Stone, J.E. and Phillips, J.C.},
booktitle={International Conference on Green Computing},
pages={317–324},
year={2010},
organization={IEEE}
}
We present an inexpensive hardware system for monitoring power usage of individual CPU hosts and externally attached GPUs in HPC clusters and the software stack for integrating the power usage data streamed in real-time by the power monitoring hardware with the cluster management software tools. We introduce a measure for quantifying the overall improvement in performance-per-watt for applications that have been ported to work on the GPUs. We use the developed hardware/software infrastructure to demonstrate the overall improvement in performance-per-watt for several HPC applications implemented to work on GPUs.
February 3, 2011 by hgpu
