Automatic data-flow graph generation of MPI programs

• R. E. Silva UFRGS
• G. Pezzi UFRGS
• N. Maillard UFRGS
• T. Diverio UFRGS

Resumo

The data-flow graph (DFG) of a parallel application is frequently used to take scheduling decisions, based on the information that it models (dependencies among the tasks and volume of exchanged data). In the case of MPI-based programs, the DFG may be built at run-time by overloading the data exchange primitives. This article presents a library that enables the generation of the DFG of a MPI program, and its use to analyze the network contention on a test-application: the Linpack benchmark. It is the first step towards automatic mapping of a MPI program on a distributed architecture.

Palavras-chave: Processor scheduling, Libraries, Parallel programming, Benchmark testing, Runtime, Yarn, Application software, Informatics, Parallel processing, Java