TY - GEN
T1 - Parallelizing the QUDA library for multi-GPU calculations in lattice quantum chromodynamics
AU - Babich, Ronald
AU - Clark, Michael A.
AU - Joó, Bálint
PY - 2010
Y1 - 2010
N2 - Graphics Processing Units (GPUs) are having a transformational effect on numerical lattice quantum chromo-dynamics (LQCD) calculations of importance in nuclear and particle physics. The QUDA library provides a package of mixed precision sparse matrix linear solvers for LQCD applications, supporting single GPUs based on NVIDIA's Compute Unified Device Architecture (CUDA). This library, interfaced to the QDP++/Chroma framework for LQCD calculations, is currently in production use on the"9g" cluster at the Jefferson Laboratory, enabling unprecedented price/performance for a range of problems in LQCD. Nevertheless, memory constraints on current GPU devices limit the problem sizes that can be tackled. In this contribution we describe the parallelization of the QUDA library onto multiple GPUs using MPI, including strategies for the overlapping of communication and computation. We report on both weak and strong scaling for up to 32 GPUs interconnected by InfiniBand, on which we sustain in excess of 4 Tflops.
AB - Graphics Processing Units (GPUs) are having a transformational effect on numerical lattice quantum chromo-dynamics (LQCD) calculations of importance in nuclear and particle physics. The QUDA library provides a package of mixed precision sparse matrix linear solvers for LQCD applications, supporting single GPUs based on NVIDIA's Compute Unified Device Architecture (CUDA). This library, interfaced to the QDP++/Chroma framework for LQCD calculations, is currently in production use on the"9g" cluster at the Jefferson Laboratory, enabling unprecedented price/performance for a range of problems in LQCD. Nevertheless, memory constraints on current GPU devices limit the problem sizes that can be tackled. In this contribution we describe the parallelization of the QUDA library onto multiple GPUs using MPI, including strategies for the overlapping of communication and computation. We report on both weak and strong scaling for up to 32 GPUs interconnected by InfiniBand, on which we sustain in excess of 4 Tflops.
UR - http://www.scopus.com/inward/record.url?scp=78650816103&partnerID=8YFLogxK
U2 - 10.1109/SC.2010.40
DO - 10.1109/SC.2010.40
M3 - Conference contribution
AN - SCOPUS:78650816103
SN - 9781424475575
T3 - 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2010
BT - 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2010
T2 - 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2010
Y2 - 13 November 2010 through 19 November 2010
ER -