We review our work done to optimize the staggered conjugate gradient (CG) algorithm in the MILC code for use with the Intel Knights Landing (KNL) architecture. KNL is the second generation Intel Xeon Phi processor. It is capable of massive thread parallelism, data parallelism, and high on-board memory bandwidth and is being adopted in supercomputing centers for scientific research. The CG solver consumes the majority of time in production running, so we have spent most of our effort on it. We compare performance of an MPI+OpenMP baseline version of the MILC code with a version incorporating the QPhiX staggered CG solver, for both one-node and multi-node runs.
|Original language||English (US)|
|Journal||Proceedings of Science|
|State||Published - 2016|
|Event||34th Annual International Symposium on Lattice Field Theory, LATTICE 2016 - Southampton, United Kingdom|
Duration: Jul 24 2016 → Jul 30 2016
ASJC Scopus subject areas