Lessons learned from implementing BSP

被引：0

作者：

Hill, JMD ^{[1
]}

Skillicorn, DB

机构：

[1] Univ Oxford, Comp Lab, Oxford OX1 2JD, England

[2] Queens Univ, Dept Comp & Informat Sci, Kingston, ON K7L 3N6, Canada

来源：

HIGH-PERFORMANCE COMPUTING AND NETWORKING | 1997年 / 1225卷

关键词：

D O I：

暂无

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

We focus on two criticisms of Bulk Synchronous Parallelism (BSP): that delaying communication until specific points in a program causes poor performance, and that frequent barrier synchronisations are too expensive for high-performance parallel computing. We show that these criticisms are misguided, not just about BSP but about parallel programming in general, because they are based on misconceptions about the origins of poor performance. The main implication for parallel programming is that higher levels of abstraction do not only make software construction easier-they also make high-performance implementation easier.

引用

页码：762 / 771

页数：10