Performance characterization of irregular I/O at the extreme scale
Abstract
Here, this paper reports our experience with irregular I/O and describes lessons learned when running applications with such I/O on supercomputers at the extreme scale. Specifically, we study how irregularities in I/O patterns (i.e., irregular amount of data written per process at each I/O step) in scientific simulations can cause increasing I/O times and substantial loss in scalability. To this end, we quantify the impact of irregular I/O patterns on the I/O performance of scientific applications at the extreme scale by statistically modeling the irregular I/O behavior of two scientific applications: the Monte Carlo application QMCPack and the adaptive mesh refinement application ENZO. For our testing, we feed our model into I/O kernels of two well-known I/O data models (i.e., ADIOS and HDF) to measure the performance of the two applications’ I/O under different I/O settings. Empirically, we show how the growing data sizes and the irregular I/O patterns in these applications are both relevant factors impacting performance.
- Authors:
-
- Univ. of Delaware, Newark, DE (United States)
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Publication Date:
- Research Org.:
- Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
- Sponsoring Org.:
- USDOE Office of Science (SC), Basic Energy Sciences (BES); National Science Foundation (NSF)
- OSTI Identifier:
- 1559754
- Alternate Identifier(s):
- OSTI ID: 1251771
- Grant/Contract Number:
- AC05-00OR22725; CCF 1318445; AAC05-00OR22725
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Parallel Computing
- Additional Journal Information:
- Journal Volume: 51; Journal Issue: C; Journal ID: ISSN 0167-8191
- Publisher:
- Elsevier
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; Exascale; Irregular I/O; QMCPack; ENZO; ADIOS; HDF5
Citation Formats
Herbein, Stephen, McDaniel, Sean, Podhorszki, Norbert, Logan, Jeremy S., Klasky, Scott A., and Taufer, Michela. Performance characterization of irregular I/O at the extreme scale. United States: N. p., 2015.
Web. doi:10.1016/j.parco.2015.10.009.
Herbein, Stephen, McDaniel, Sean, Podhorszki, Norbert, Logan, Jeremy S., Klasky, Scott A., & Taufer, Michela. Performance characterization of irregular I/O at the extreme scale. United States. https://doi.org/10.1016/j.parco.2015.10.009
Herbein, Stephen, McDaniel, Sean, Podhorszki, Norbert, Logan, Jeremy S., Klasky, Scott A., and Taufer, Michela. Sat .
"Performance characterization of irregular I/O at the extreme scale". United States. https://doi.org/10.1016/j.parco.2015.10.009. https://www.osti.gov/servlets/purl/1559754.
@article{osti_1559754,
title = {Performance characterization of irregular I/O at the extreme scale},
author = {Herbein, Stephen and McDaniel, Sean and Podhorszki, Norbert and Logan, Jeremy S. and Klasky, Scott A. and Taufer, Michela},
abstractNote = {Here, this paper reports our experience with irregular I/O and describes lessons learned when running applications with such I/O on supercomputers at the extreme scale. Specifically, we study how irregularities in I/O patterns (i.e., irregular amount of data written per process at each I/O step) in scientific simulations can cause increasing I/O times and substantial loss in scalability. To this end, we quantify the impact of irregular I/O patterns on the I/O performance of scientific applications at the extreme scale by statistically modeling the irregular I/O behavior of two scientific applications: the Monte Carlo application QMCPack and the adaptive mesh refinement application ENZO. For our testing, we feed our model into I/O kernels of two well-known I/O data models (i.e., ADIOS and HDF) to measure the performance of the two applications’ I/O under different I/O settings. Empirically, we show how the growing data sizes and the irregular I/O patterns in these applications are both relevant factors impacting performance.},
doi = {10.1016/j.parco.2015.10.009},
journal = {Parallel Computing},
number = C,
volume = 51,
place = {United States},
year = {Sat Oct 24 00:00:00 EDT 2015},
month = {Sat Oct 24 00:00:00 EDT 2015}
}
Web of Science