Cache Line Sharing and Communication in ECP Proxy Applications

摘要: Scientific computing codes rely on efficient parallelization to achieve performance. This parallel efficiency is reduced by factors such as communication, serialization, and data sharing. In this work, we examine interactions between OpenMP threads in the context of a Chip-multiprocessor (CMP). We first analyze cache line sharing observe how often multiple are accessing same data. then look at producer-consumer write-invalidation these threads. These implemented with coherence operations demonstrate interference find that none studied show prohibitive amounts communication many follow simple patterns. Our work discovers opportunities increase analyzed provides motivating for research into CMP design.