skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: StoreRush: An Application-Level Approach to Harvesting Idle Storage in a Best Effort Environment

Abstract

For a production HPC system where storage devices are shared between multiple applications and managed in a best effort manner, contention is often a major problem leading to some storage devices being more loaded than others and causing a significant reduction in I/O throughput. In this paper, we describe our latest efforts StoreRush to resolve this practical issue at the application level without requiring modification to the file and storage system. The proposed scheme uses a two-level messaging system to harvest idle storage via re-routing I/O requests to a less congested storage location so that write performance is improved while limiting the impact on read by throttling re-routing if deemed too much. An analytical model is derived to guide the setup of optimal throttling factor. The proposed scheme is verified against production applications Pixie3D, XGC1 and QMCPack during production windows, which very well demonstrated the effectiveness (e.g., up to 1.8x improvement in write) and scalability of our approach (up to 131,072 cores).

Authors:
 [1];  [2];  [2];  [2];  [2];  [2];  [3];  [4]
  1. New Jersey Institute of Technology, Newark, NJ (United States)
  2. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
  3. Stony Brook Univ., NY (United States)
  4. Temple Univ., Philadelphia, PA (United States)
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1567443
Resource Type:
Accepted Manuscript
Journal Name:
Procedia Computer Science
Additional Journal Information:
Journal Volume: 108; Journal Issue: C; Journal ID: ISSN 1877-0509
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Computer Science; High Performance Computing; Storage; I/O

Citation Formats

Liu, Qing, Podhorszki, Norbert, Choi, Jong, Logan, Jeremy, Wolf, Matt, Klasky, Scott, Kurc, Tahsin, and He, Xubin. StoreRush: An Application-Level Approach to Harvesting Idle Storage in a Best Effort Environment. United States: N. p., 2017. Web. doi:10.1016/j.procs.2017.05.005.
Liu, Qing, Podhorszki, Norbert, Choi, Jong, Logan, Jeremy, Wolf, Matt, Klasky, Scott, Kurc, Tahsin, & He, Xubin. StoreRush: An Application-Level Approach to Harvesting Idle Storage in a Best Effort Environment. United States. doi:10.1016/j.procs.2017.05.005.
Liu, Qing, Podhorszki, Norbert, Choi, Jong, Logan, Jeremy, Wolf, Matt, Klasky, Scott, Kurc, Tahsin, and He, Xubin. Fri . "StoreRush: An Application-Level Approach to Harvesting Idle Storage in a Best Effort Environment". United States. doi:10.1016/j.procs.2017.05.005. https://www.osti.gov/servlets/purl/1567443.
@article{osti_1567443,
title = {StoreRush: An Application-Level Approach to Harvesting Idle Storage in a Best Effort Environment},
author = {Liu, Qing and Podhorszki, Norbert and Choi, Jong and Logan, Jeremy and Wolf, Matt and Klasky, Scott and Kurc, Tahsin and He, Xubin},
abstractNote = {For a production HPC system where storage devices are shared between multiple applications and managed in a best effort manner, contention is often a major problem leading to some storage devices being more loaded than others and causing a significant reduction in I/O throughput. In this paper, we describe our latest efforts StoreRush to resolve this practical issue at the application level without requiring modification to the file and storage system. The proposed scheme uses a two-level messaging system to harvest idle storage via re-routing I/O requests to a less congested storage location so that write performance is improved while limiting the impact on read by throttling re-routing if deemed too much. An analytical model is derived to guide the setup of optimal throttling factor. The proposed scheme is verified against production applications Pixie3D, XGC1 and QMCPack during production windows, which very well demonstrated the effectiveness (e.g., up to 1.8x improvement in write) and scalability of our approach (up to 131,072 cores).},
doi = {10.1016/j.procs.2017.05.005},
journal = {Procedia Computer Science},
number = C,
volume = 108,
place = {United States},
year = {2017},
month = {6}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 1 work
Citation information provided by
Web of Science

Save / Share: