Performance and library issues for mathematical software on high performance computers
This paper discusses some of the fundamental issues facing designers of mathematical software libraries for medium scale parallel processors such as the CRAY X-MP-4 and the Denelcor HEP. We discuss the problems that arise with performance and demonstrate that it may be appropriate to exploit parallelism at all levels of the program, not just at the highest level. We give performance measurements indicating the efficiency of a linear algebra library written in terms of a few high level modules. These modules chosen at the matrix vector level extend the concept of the BLAS and provide enough computational granularity to allow efficient implementations on a wide variety of architectures. Only three modules must be recoded for efficiency in order to transport the library to various machines. We report experience on machines as diverse as the CRAY X-MP and the Denelcor HEP. Finally, we report on some special algorithms for the HEP which take advantage of the fine grain parallelism capabilities.
- Research Organization:
- Argonne National Lab., IL (USA)
- DOE Contract Number:
- W-31-109-ENG-38
- OSTI ID:
- 5746331
- Report Number(s):
- CONF-8411140-4; ON: DE85010263
- Resource Relation:
- Conference: ARO workshop on new computing environments: parallel, vector and systolic, Stanford, CA, USA, 7 Nov 1984; Other Information: Portions of this document are illegible in microfiche products
- Country of Publication:
- United States
- Language:
- English
Similar Records
Signal processing algorithms on parallel architectures: A performance update
Extreme-Scale Algorithms & Software Resilience (EASIR) Architecture-Aware Algorithms for Scalable Performance and Resilience on Heterogeneous Architectures