Experimental Study of Compiler Techniques for NUMA Machines

This study1 explores the applicability of fully automatic parallelizing techniques for parallel computers. In this study, we capitalize on a variety of traditional compiling techniques as well as new techniques developed specifically for distributed memory architectures. Combining these traditional and new techniques, we conducted experiments with several benchmark programs on the Cray T3D.

• Type: Conference
• Year: 1998
• Where: IPPS
• Authors: Yunheung Paek, David A. Padua
• Keywords: Automatic Parallelizing Techniques, Distributed And Parallel Computing, IPPS 1998, Parallel Computers, Traditional Compiling Techniques