๐”– Bobbio Scriptorium
โœฆ   LIBER   โœฆ

Portability, predictability and performance for parallel computing: BSP in practice

โœ Scribed by Reed, Joy; Parrott, Kevin; Lanfear, Tim

Publisher
John Wiley and Sons
Year
1996
Tongue
English
Weight
783 KB
Volume
8
Category
Article
ISSN
1040-3108

No coin nor oath required. For personal study only.

โœฆ Synopsis

We report on practical experience using the Oxford BSP Library to parallelize a large electromagnetic code, the British Aerospace finite-difference time-domain code EMMA TFD3D. The Oxford BSP Library is one of the first realizations of the Bulk Synchronous Parallel computational model to be targeted at numerically intensive scientific (typically Fortran) computing. The BAe EMMA code is one of the first large-scale applications to be parallelized using this library, and it is an important demonstration of the cost effectiveness of the BSP approach. We illustrate how BSP cost-modelling techniques can be used to predict and optimize performance for single-source programs across different parallel platforms. We provide predicted and observed performance figures for an industrial-strength, single-source parallel code for a variety of real parallel architectures: shared memory multiprocessors, workstation clusters and massively parallel platforms.

๐Ÿ“œ SIMILAR VOLUMES

An Effective and Practical Performance P
An Effective and Practical Performance Prediction Model for Parallel Computing on Nondedicated Heterogeneous NOW
โœ Yong Yang; Xiaodong Zhang; Yongsheng Song ๐Ÿ“‚ Article ๐Ÿ“… 1996 ๐Ÿ› Elsevier Science ๐ŸŒ English โš– 487 KB

riously hard because of the heterogeneity and non-dedicated system effects of the underlying system. In this paper, we focus on developing a practical performance prediction methodology for such a network of workstations. Quite a few researchers [1,3,7,8,9,12] have contributed useful results to the

An Effective and Practicle Performance P
An Effective and Practicle Performance Prediction Model for Parallel Computing on Nondedicated Heterogeneous NOW: Volume38,Number 1 (1996), pages 63โ€“80
โœ Yong Yan; Xiaodong Zhang; Yongsheng Song ๐Ÿ“‚ Article ๐Ÿ“… 1997 ๐Ÿ› Elsevier Science ๐ŸŒ English โš– 10 KB

The source of the owner workload model in Sections 3.2.3 and 2.1 was not properly attributed. The discrete time model and model description are directly borrowed from Leutenegger and Sun [1]. A minor modification to the model of Leutenegger and Sun was made to incorporate the processing power and ow