Compiler optimizations for Fortran D on MIMD distributed-memory machines
Proceedings of the 1991 ACM/IEEE conference on Supercomputing
Dynamic data distributions in Vienna Fortran
Proceedings of the 1993 ACM/IEEE conference on Supercomputing
Parallel Programming with Polaris
Computer
Compiling Communication-Efficient Programs for Massively Parallel Machines
IEEE Transactions on Parallel and Distributed Systems
An Advanced Compiler Framework for Non-Cache-Coherent Multiprocessors
IEEE Transactions on Parallel and Distributed Systems
Experimental Study of Compiler Techniques for NUMA Machines
IPPS '98 Proceedings of the 12th. International Parallel Processing Symposium on International Parallel Processing Symposium
Optimizing Data Scheduling on Processor-In-Memory Arrays
IPPS '98 Proceedings of the 12th. International Parallel Processing Symposium on International Parallel Processing Symposium
| Hi-index | 0.00 |
The Polaris restructurer transforms conventional Fortran programs into parallel form for various types of multiprocessor systems. This paper presents the results of a study on strategies to improve the effectiveness of Polaris' techniques for distributed-memory multiprocessors. Our study, which is based on the hand analysis of MDG and TRFD from the Perfect Benchmarks and TOMCATV and SWIM from SPEC benchmarks, identified three techniques that are important for improving communication optimization. Their application produces almost perfect speedups for the four programs on the Cray T3D.