J4 ›› 2005, Vol. 27 ›› Issue (9): 105-107.
• 论文 • 上一篇 下一篇
曾丽芳 杨学军
出版日期:
发布日期:
Online:
Published:
摘要：
FORALL结构是FORTRAN 95的一种语法，在编译器中高效地实现FORALL结构是一项富有挑战性的工作，局部性优化对其高效实现尤其重要。本文介绍作者在G95编译器中实现FOR ALL结构时用到的两种局部性优化方法——临时空间合并和嵌套循环排序。实验结果表明，局部性优化对提高FORALL结构的性能非常有效。对某类FORALL结构，与Intel的EFC 编译器相比，我们的实现方法能提高30％的性能。
关键词: 局部性优化 FORALL结构 临时空间合并 嵌套循环排序
Abstract:
The FORALL construct is a Fortran 95 syntax, the effective implementation of the FORALL construct is full of challenge. Especially, the locality optim ization is very important for the compiler to generate efficient code when implementing the FORALL construct. In this paper, we introduce two locality o ptimizing methods, Temporary Space Group and Nested Loop Sort, to implement the FORALL construct in the G 95 compiler. The experimental results show that locality optimization is very efficient to gain speed benefit. Compared with the Intel EFC compiler, our implementation gains a performance improvemet up to 30% for some kind of applications.
Key words: （locality optimization, FORALL construct, temporary space group, nested loop sort）
曾丽芳 杨学军. FORALL实现过程中的局部性优化[J]. J4, 2005, 27(9): 105-107.
0 / / 推荐
导出引用管理器 EndNote|Ris|BibTeX
链接本文: http://joces.nudt.edu.cn/CN/
http://joces.nudt.edu.cn/CN/Y2005/V27/I9/105
