Brown, David C.
Looft, Fred J.
One of the most challenging procedures in parallel programming is to convert an algorithm written in sequential code to an efficient parallel version. This paper discusses a program monitoring tool, Graphical Analysis of Clustered Systems (GACS), that was created to assist in analyzing a parallel program's execution by displaying its trace in a graphical interface. The goal of GACS is to help programmers to determine where the bottlenecks are in a parallel application and, subsequently, to allow them to improve the performance of their code. GACS accomplishes this by allowing the developer to see when certain code and message passing functions occur on each of the processors relative to similar functions on the other processors.
Worcester Polytechnic Institute
Major Qualifying Project
Access to this report is limited to members of the WPI community. Please contact a project advisor or their department to request access
Restricted-WPI community only
Electrical and Computer Engineering