Open AccessParallel Storage Devices Profiling with SeRRa
Author(s) -
V.S. Machado,
Francieli Za Boito,
Rodrigo Kassick,
Jean Luca Bez,
Philippe O. A. Navaux,
Yves Denneulin
Publication year - 2015
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/wperformance.2015.10401
Subject(s) - profiling (computer programming) , computer science , parallel computing , throughput , execution time , operating system , wireless
This work presents the parallel storage device profiling tool SeRRa. Our tool obtains the sequential to random throughput ratio for reads and writes of different sizes on storage devices. In order to provide this information efficiently, SeRRa employs benchmarks to obtain the values for only a subset of the parameter space and estimates the remaining values through linear models. The MPI parallelization of SeRRa presented in this paper allows for faster profiling. Our results show that our parallel SeRRa provides profiles up to 8:7 times faster than the sequential implementation, up to 895 times faster than the originally required time (without SeRRa).