Comparative performance evaluation of cache-coherent NUMA and COMA architectures

Abstract

Two interesting variations of large-scale shared-memory machines that have recently emerged are cache-coherent mmumform-memory-access machines (CC-NUMA) and cacheonly memory architectures (COMA). They both have distributed main memory and use directory-based cache coherence. Unlike CC-NUMA, however, COMA machines automatically migrate and replicate data at the main-memoty level in cache-line sized chunks. This paper compares the performance of these two classes of machines.We first present a qualitative model that shows that the relative performance is primarily determined by two factors: the relative magnitude of capacity misses versus coherence misses, and the gramrhirity of data partitions in the application.We then present quantitative results using simulation studies for eight prtraUeI applications (including all six applications from the SPLASH benchmark suite). We show that COMA's potential for performance improvement is limited to applications where data accesses by different processors are finely interleaved in memory space and, in addition, where capacity misses dominate over coherence misses. In other situations, for example where coherence misses dominate, COMA can actually perform worse than CC-NUMA due to increased miss latencies caused by its hierarchical directories. Finally, we propose a new architectural alternative, called COMA-F, that combines the advantages of both CC-NUMA and COMA.

Parallelizing Shared File I/O Operations of NVM File System for Manycore Servers

“…Then, the modi cation of this shared variable results in invalidating the cache lines including the lock variable of waiting threads and even lock holders. In particular, when a cache line is shared among threads in a NUMA environment, frequent invocations of the cache line invalidation to maintain the cache coherence become a signi cant overhead [19], [29]. In the case of SRL(tree,i), when it dynamically allocates slvs, the cache line for atomic operations can be shared.…”