Parallelization and performance of Conjugate Gradient algorithms on the Cedar hierarchical-memory multiprocessor

Ulrike Meier, Rudolf Eigenmann
1991 Proceedings of the third ACM SIGPLAN symposium on Principles and practice of parallel programming - PPOPP '91  
The conjugate gradient method is a powerful algorithm for solving well-structured sparse linear systems that arise from partial di erential equations. The broad application range makes it an interesting object for investigating novel architectures and programming systems. In this paper we analyze the computational structure of three di erent conjugate gradient schemes for solving elliptic partial di erential equations. We describe its parallel implementation on the Cedar hierarchical memory
more » ... iprocessor from both angles, explicit manual parallelization and automatic compilation. We report performance measurements taken on Cedar, which allow u s a n umber of conclusions on the Cedar architecture, the programming methodology for hierarchical computer structures, and the contrast of manual vs automatic parallelization.
doi:10.1145/109625.109644 dblp:conf/ppopp/MeierE91 fatcat:eakzchabpvht3dt5fg3njrkswa