Parallel Processing Symposium, International (1997)
Apr. 1, 1997 to Apr. 5, 1997
Pete Keleher & Chau-Wen Tseng , University of Maryland
Current parallelizing compilers for message-passing machines only support a limited class of data-parallel applications. One method for eliminating this restriction is to combine powerful shared-memory parallelizing compilers with software distributed-shared-memory (DSM) systems. We demonstrate such a system by combining the SUIF parallelizing compiler and the CVM software DSM. Innovations of the system include compiler-directed techniques that: 1) combine synchronization and parallelism information communication on parallel task invocation, 2) employ customized routines for evaluating reduction operations, and 3) select a hybrid update protocol that pre-sends data by flushing updates at barriers.For applications with sufficient granularity of parallelism, these optimizations yield very good eight processor speedups on an IBM SP-2 and DEC Alpha cluster, usually matching or exceeding the speedup of equivalent HPF and message-passing versions of each program. Flushing updates, in particular, eliminates almost all nonlocal memory misses and improves performance by 13% on average.
P. K. Tseng, "Enhancing Software DSM for Compiler-Parallelized Applications," Parallel Processing Symposium, International(IPPS), Geneva, SWITZERLAND, 1997, pp. 490.