Argonne National Laboratory

Automatic Generation of I/O Kernels for HPC Applications

TitleAutomatic Generation of I/O Kernels for HPC Applications
Publication TypeConference Paper
Year of Publication2014
AuthorsBehzad, B, Dang, H, Hariri, F, Zhang, W, Snir, M
Conference NameParallel Data Storage Workshop
Date Published11/2014
Other NumbersANL/MCS-P5213-1014
AbstractThe study of the I/O performance of a parallel application can be facilitated by the use of an I/O skeleton program – a program that generates the same I/O calls as the original application, but can be executed much faster. Such trace generators are especially important when the programs under study are proprietary or classified, and only available in binary form. In this paper, we show how to create automatically such an I/O skeleton program, by executing the target application with an instrumented I/O library, next “compressing” the resulting I/O traces into a compact C program that generates those traces.