N-Body simulation
Abstract
Presents performance optimization methods applicable to N-body simulations on Knights Landing. Discusses the optimization of arithmetic expressions, data structures, thread parallelism, and memory traffic on Knights Landing. Results demonstrate that the N-body simulation, previously optimized for parallelism on Knights Corner, achieves most of the performance improvements available with Knights Landing without any code adaptation—only a recompilation is necessary to take advantage of AVX-512.
Keywords
N-body; Strength reduction; Alignment; Vectorization; SoA; AoS; Unit-stride; OpenMP; Threads; Affinity; Tiling; MCDRAM
Get Intel Xeon Phi Processor High Performance Programming, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
