Abstract
The very nature of distributed-memory parallel architectures demands the serious consideration of interprocessor communications, by which non-local memory access can be implemented. In this paper, we propose a method for optimization that utilizes asynchronous and bulk communications. We constructed an HPF- compiler, a subset of HPF, and evaluated it on the CM5, a true distributed-memory machine using three non-trivial benchmark programs. With optimized code, there was considerable improvement in communications over non-optimized code and we had much better results than when optimizing by means of CM Fortran.
Original language | English |
---|---|
Pages | 185-189 |
Number of pages | 5 |
Publication status | Published - Jan 1 1995 |
Event | Proceedings of the 1995 Conference on Supercomputing - Barcelona, Spain Duration: Jul 3 1995 → Jul 7 1995 |
Other
Other | Proceedings of the 1995 Conference on Supercomputing |
---|---|
City | Barcelona, Spain |
Period | 7/3/95 → 7/7/95 |
All Science Journal Classification (ASJC) codes
- Computer Science(all)