Performance of parallel computers for spectral atmospheric models

Abstract

Massively parallel processing (MPP) computer systems use high-speed interconnection networks to link hundreds or thousands of RISC microprocessors. With each microprocessor having a peak performance of 100 Mflops/sec or more, there is at least the possibility of achieving very high performance. However, the question of exactly how to achieve this performance remains unanswered. MPP systems and vector multiprocessors require very different coding styles. Different MPP systems have widely varying architectures and performance characteristics. For most problems, a range of different parallel algorithms is possible, again with varying performance characteristics. In this paper, we provide a detailed, fair evaluation of MPP performance for a weather and climate modeling application. Using a specially designed spectral transform code, we study performance on three different MPP systems: Intel Paragon, IBM SP2, and Cray T3D. We take great care to control for performance differences due to varying algorithmic characteristics. The results yield insights into MPP performance characteristics, parallel spectral transform algorithms, and coding style for MPP systems. We conclude that it is possible to construct parallel models that achieve multi-Gflop/sec performance on a range of MPPs if the models are constructed to allow run-time selection of appropriate algorithms.