S. Jenks (USA)
Parallelism, Multithreading, Migration, Interconnection
The balance between CPU speed and interconnection net work throughput in distributed memory parallel computers varies with each generation of systems, but the trend is that CPUs are gaining performance faster than the interconnec tion networks. This means that remote data accesses are be coming more expensive relative to local accesses in terms of CPU cycles. Therefore, remote memory access mecha nisms that were suited to a previous generation of parallel machines may be less appropriate for current clusters. This research evaluates a multithreaded program ming paradigm with cached remote memory accesses and thread migration to exploit array locality on a cluster with Myrinet. The approach, called Nomadic Threads, was orig inally developed for the CM5, but has been adapted to use MPI on Linux clusters. The results show that the current surfeit of CPU power vs. network throughput dramatically changes scaling characteristics of some programs while others behave much as they did on the decade-old CM5.
Important Links:
Go Back
