Optimization of the Brillouin operator on the KNL architecture

Abstract

Experiences with optimizing the matrix-times-vector application of the Brillouin operator on the Intel KNL processor are reported. Without adjustments to the memory layout, performance figures of 360 Gflop/s in single and 270 Gflop/s in double precision are observed. This is with Nc = 3 colors, Nv = 12 right-hand-sides, Nthr = 256 threads, on lattices of size 323 × 64, using exclusively OMP pragmas. Interestingly, the same routine performs quite well on Intel Core i7 architectures, too. Some observations on the much harderWilson fermion matrix-times-vector optimization problem are added.