Title :
Automated scalability analysis of message-passing parallel programs
Author :
Sarukkai, Sekhar R. ; Mehra, Pankaj ; Block, Robert J.
Author_Institution :
NASA Ames Res. Center, Moffett Field, CA, USA
Abstract :
Scalability analysis, which characterizes large-scale performance, is indispensable for parallel-program performance debugging. To assist developers in this usually difficult process, we´ve developed a methodology and a toolkit that provide automatic, fast and accurate scalability analysis for a class of deterministic message-passing scientific applications. Modeling Kernel, our scalability analysis toolkit, generates a model based on a program´s parse tree, which represents the program´s syntactic structure. We have successfully demonstrated our approach by automatically characterizing the scalability of several scientific applications that run on Intel´s iPSC/860 and Paragon supercomputers. To characterize large-scale performance, this scalability analysis toolkit constructs augmented parse trees (APTs). APTs combine two key data structures: annotated parse trees and communication phase graphs. By parsing the APT, the toolkit supports simulation, abstract interpretation and complexity analysis
Keywords :
digital simulation; message passing; parallel programming; program debugging; program diagnostics; software performance evaluation; software tools; tree data structures; Intel Paragon supercomputers; Intel iPSC/860; Modeling Kernel; abstract interpretation; annotated parse trees; augmented parse trees; automated scalability analysis; communication phase graphs; complexity analysis; data structures; deterministic scientific applications; large-scale performance; message-passing parallel programs; parallel-program performance debugging; program syntactic structure; scalability analysis toolkit; simulation; Analytical models; Computational fluid dynamics; Computer aided instruction; Large-scale systems; Monitoring; Performance analysis; Programming profession; Scalability; Tree graphs; Visualization;
Journal_Title :
Parallel & Distributed Technology: Systems & Applications, IEEE