Oversubscription on multicore processors

Costin Iancu, Steven Hofmeyr, Filip Blagojevic, Yili Zheng
2010 2010 IEEE International Symposium on Parallel & Distributed Processing (IPDPS)  
Existing multicore systems already provide deep levels of thread parallelism. Hybrid programming models and composability of parallel libraries are very active areas of research within the scientific programming community. As more applications and libraries become parallel, scenarios where multiple threads compete for a core are unavoidable. In this paper we evaluate the impact of task oversubscription on the performance of MPI, OpenMP and UPC implementations of the NAS Parallel Benchmarks on
more » ... A and NUMA multisocket architectures. We evaluate explicit thread affinity management against the default Linux load balancing and discuss sharing and partitioning system management techniques. Our results indicate that oversubscription provides beneficial effects for applications running in competitive environments. Sharing all the available cores between applications provides better throughput than explicit partitioning. Modest levels of oversubscription improve system throughput by 27% and provide better performance isolation of applications from their co-runners: best overall throughput is always observed when applications share cores and each is executed with multiple threads per core. Rather than "resource" symbiosis, our results indicate that the determining behavioral factor when applications share a system is the granularity of the synchronization operations.
doi:10.1109/ipdps.2010.5470434 dblp:conf/ipps/IancuHBZ10 fatcat:vomcoi2h7fg43cnu4jxyhqvjbe