Evaluating The Performance of Non-Blocking Synchronisation on Modern Shared-Memory Multiprocessors

Parallel programs running on shared memory multiprocessors coordinate via shared data objects/structures. To ensure the consistency of the shared data structures, programs typically rely on some forms of software synchronisations. Unfortunately typical software synchronisation mechanisms usually result in poor performance because they produce large amounts of memory and interconnection network contention and, more signi cantly, because they produce convoy e ects that degrade signi cantly in multiprogramming environments: if one process holding a lock is preempted, other processes on di erent processors waiting for the lock will not be able to proceed. Researchers have introduced non-blocking synchronisation to address the above problems. However, its performance implications are not well understood on modern systems or on real applications. In this paper we study the impact of the non-blocking synchronisation on parallel applications running on top of a modern, 64 processor, cache-coherent, shared memory multiprocessor system: the SGI Origin 2000. In addition to the performance results on a modern system, we investigate the key synchronisation schemes that are used in multiprocessor applications and their e cient transformation to non-blocking ones.