Java and Scientific Programming: Portable Browsers for Performance Programming
Michał Cierniak, Suresh Srinivas · 2007
We present jCITE, a performance tuning tool for scientific applications. By combining the static information produced by the compiler with the profile data from real program execution, jCITE can be used to quickly understand the performance bottlenecks. The compiler information allows great understanding of what optimizations have been performed. The user can also find out which optimization have not been applied and why. Platform independence makes Java the ideal implementation platform for our tool. SGI users can have the same performance analysis tool on all platforms. You can run jCITE on an SGI O2 computer to optimize code for that machine, or you can run jCITE from within Netscape on a SPARCstation to analyze performance of a Cray application. In our experiments we were able to significantly speed up some SPEC95 applications in a few days or even a few hours without any prior knowledge of those applications. 1 Introduction – the need for performance programming for Scientific Computing Performance is at the heart of scientific and engineering computing. Tuning a critical inner loop could lead to drastic reduction in the running time of a large scientific or engineering application. This paper is about portable performance programming tools for the scientific and engineering programmer. Larry Carter and Bowen Alpern define performance programming as “programming when a 30 % improvement in speed