Experimental results for vector processing on the Alliant FX/8
The Alliant FX/8 multiprocessor implements several high-speed computation ideas in software and hardware. Each of the 8 computational elements (CEs) has vector capabilities and multiprocessor support. Generally, the FX/8 delivers its highest processing rates when executing vector loops concurrently. In this paper, we present extensive empirical performance results for vector processing on the FX/8. The vector kernels of the LANL BMK8a1 benchmark are used in the experiments. We execute each kernel on 1 and 8 CEs and show the measured execution rate (in MFLOPS) as a functtion of vector length. We assess the performance of 1 CE as a vector processor by finding the vector lengths where vector processing exceeds that of scalar processing and calculating Hockney's n/sub 1/2/. For 8 CEs, we give upper/lower bounds on the achieved speedups and on the multiprocessing overhead. We also show the speedup variation as the number of CEs increases from 2 to 8. 24 refs., 84 figs., 7 tabs.
- Research Organization:
- Illinois Univ., Urbana (USA). Center for Supercomputing Research and Development
- DOE Contract Number:
- FG02-85ER25001
- OSTI ID:
- 7204176
- Report Number(s):
- DOE/ER/25001-15; CSRD-539; ON: DE87002095
- Country of Publication:
- United States
- Language:
- English
Similar Records
Performance of the Lawrence Livermore National Lab benchmark on the Alliant FX/8
Performance of a Benchmark Implementation of the Van Slyke and Wets Algorithm for Stochastic Programs on the Alliant FX/8