Code coverage, performance approximation and automatic recognition of idioms in scientific applications

Basic data flow patterns which we call idioms, such as stream, transpose, reduction, random access and stencil, are common in scientific numerical applications. We hypothesize that a small number of idioms can cover most programming constructs that dominate the execution time of scientific codes and can be used to approximate the application performance. In this paper, we start with a manual analysis of code coverage on the NAS Parallel Benchmark (NPB) and find that five idioms suffice to cover 100% of the NPB codes. We then compare the performance of our idiom benchmarks and their corresponding instances in different NPB codes on two different platforms and find that they differ by about 30%. To check the hypotheses with real applications further, we propose an automatic idioms recognition method, implement the method basing on the open source compiler Open64, and verify the prototype system with the previous manual analysis results. Categories and Subject Descriptors C.4 [Computer S...

• Type: Conference
• Year: 2008
• Where: HPDC
• Authors: Jiahua He, Allan Snavely, Rob F. Van der Wijngaart, Michael A. Frumkin
• Keywords: Distributed And Parallel Computing, HPDC 2008, Idioms, NPB Codes, Performance