TY - GEN
T1 - High-performance lattice QCD for multi-core based parallel systems using a cache-friendly hybrid threaded-MPI approach
AU - Smelyanskiy, Mikhail
AU - Vaidyanathan, Karthikeyan
AU - Choi, Jee
AU - Joó, Bálint
AU - Chhugani, Jatin
AU - Clark, Michael A.
AU - Dubey, Pradeep
PY - 2011
Y1 - 2011
N2 - Lattice Quantum Chromo-dynamics (LQCD) is a computationally challenging problem that solves the discretized Dirac equation in the presence of an SU(3) gauge field. Its key operation is a matrixvector product, known as the Dslash operator. We have developed a novel multicore architecture-friendly implementation of the Wilson-Dslash operator which delivers 75 Gflops (single-precision) on an Intel® Xeon® Processor X5680 achieving 60% computational efficiency for datasets that fit in the last-level cache. For datasets larger than the last-level cache, this performance drops to 50 Gflops. Our performance is 2-3X higher than a well-known implementation from the Chroma software suite when running on the same hardware platform. The novel implementation of LQCD reported in this paper is based on recently published the 3.5D spatial and 4.5D temporal tiling schemes. Both blocking schemes significantly reduce LQCD external memory bandwidth requirements, delivering a more compute-bound implementation. The performance advantage of our schemes will become more significant as the gap between compute flops and external memory bandwidth continues to grow. We demonstrate very good cluster-level scalability of our implementation: for a lattice of 323×256 sites, we achieve over 4 Tflops when strong-scaled to a 128 node system (1536 cores total). For the same lattice size, a full Conjugate Gradients Wilson-Dslash operator, achieves 2.95 Tflops.
AB - Lattice Quantum Chromo-dynamics (LQCD) is a computationally challenging problem that solves the discretized Dirac equation in the presence of an SU(3) gauge field. Its key operation is a matrixvector product, known as the Dslash operator. We have developed a novel multicore architecture-friendly implementation of the Wilson-Dslash operator which delivers 75 Gflops (single-precision) on an Intel® Xeon® Processor X5680 achieving 60% computational efficiency for datasets that fit in the last-level cache. For datasets larger than the last-level cache, this performance drops to 50 Gflops. Our performance is 2-3X higher than a well-known implementation from the Chroma software suite when running on the same hardware platform. The novel implementation of LQCD reported in this paper is based on recently published the 3.5D spatial and 4.5D temporal tiling schemes. Both blocking schemes significantly reduce LQCD external memory bandwidth requirements, delivering a more compute-bound implementation. The performance advantage of our schemes will become more significant as the gap between compute flops and external memory bandwidth continues to grow. We demonstrate very good cluster-level scalability of our implementation: for a lattice of 323×256 sites, we achieve over 4 Tflops when strong-scaled to a 128 node system (1536 cores total). For the same lattice size, a full Conjugate Gradients Wilson-Dslash operator, achieves 2.95 Tflops.
UR - http://www.scopus.com/inward/record.url?scp=83155182851&partnerID=8YFLogxK
U2 - 10.1145/2063384.2063477
DO - 10.1145/2063384.2063477
M3 - Conference contribution
AN - SCOPUS:83155182851
SN - 9781450307710
T3 - Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis
BT - Proceedings of 2011 SC - International Conference for High Performance Computing, Networking, Storage and Analysis
T2 - 2011 International Conference for High Performance Computing, Networking, Storage and Analysis, SC11
Y2 - 12 November 2011 through 18 November 2011
ER -