%0 Conference Paper
%A Mao, Guoyong
%A Böhme, David
%A Hermanns, Marc-André
%A Geimer, Markus
%A Lorenz, Daniel
%A Wolf, Felix
%T Catching Idlers with Ease: A Lightweight Wait-State Profiler for MPI Programs
%C New York, NY, USA
%I ACM Press
%M FZJ-2014-05436
%@ 978-1-4503-2875-3
%P 103
%D 2014
%< Proceedings of the 21st European MPI Users' Group Meeting
%X Load imbalance usually introduces wait states into the execution of parallel programs. Being able to identify and quantify wait states is therefore essential for the diagnosis and remediation of this phenomenon. An established method of detecting wait states is to generate event traces and compare relevant timestamps across process boundaries. However, large trace volumes usually prevent the analysis of longer execution periods. In this paper, we present an extremely lightweight wait-state profiler which does not rely on traces that can be used to estimate wait states in MPI codes with arbitrarily long runtimes. The profiler combines scalability with portability and low overhead.
%B EuroMPI/ASIA 21st European MPI Users' Group Meeting
%C 9 Sep 2014 - 12 Sep 2014, Kyoto (Japan)
Y2 9 Sep 2014 - 12 Sep 2014
M2 Kyoto, Japan
%F PUB:(DE-HGF)8 ; PUB:(DE-HGF)7
%9 Contribution to a conference proceedingsContribution to a book
%R 10.1145/2642769.2642783
%U https://juser.fz-juelich.de/record/171880