The Community for Technology Leaders
2009 International Conference on Parallel Processing Workshops (2009)
Vienna, Austria
Sept. 22, 2009 to Sept. 25, 2009
ISSN: 1530-2016
ISBN: 978-0-7695-3803-7
pp: 105-112
ABSTRACT
We present a collection of techniques for exploiting latent I/O asynchrony which can substantially improve performance in data-intensive parallel applications. Latent asynchrony refers to an application's tolerance for decoupling ancillary operations from its core computation, and is a property of HPC codes not fully explored by current HPC I/O systems. Decoupling operations such as buffering and staging, reorganization, and format conversion in space and in time from core codes can shorten I/O phases, preserving valuable MPP compute cycles. We describe in this paper DataTaps, IOgraphs, and Metabots, three tools which allow HPC developers to implement decoupled I/O operations. Using these tools, asynchrony can be exploited by data generators which overlap computation with communication, and by data consumers that perform data conversion and reorganization out-of-band and on-demand. In the context of a data-intensive fusion simulation, we show that exploiting latent asynchrony through decoupling of operations can provide significant performance benefits.
INDEX TERMS
CITATION

M. Wolf et al., "Exploiting Latent I/O Asynchrony in Petascale Science Applications," 2009 International Conference on Parallel Processing Workshops(ICPPW), Vienna, Austria, 2009, pp. 105-112.
doi:10.1109/ICPPW.2009.67
83 ms
(Ver 3.3 (11022016))