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

Title: Rootless Operations for MPI (RLO) v0.1

Abstract

Flexible communication routines that enable high-performance computing (HPC) applications to operate with less synchronization are a great benefit to application algorithm creation and developer productivity. Further, communication operations that reduce synchronization in HPC applications while continuing to scale well are critical to application performance in the exascale era. We have implemented mechanisms that allow an application to scalably execute bcast/iAllReduce from any MPI rank without specifying the origin of the message in advance, demonstrating application development flexibility, highly scalable operation, and reduced application synchronization.

Developers:
 [1];  [1]
  1. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Release Date:
Project Type:
Open Source, Publicly Available Repository
Software Type:
Scientific
Licenses:
BSD 3-clause "New" or "Revised" License
Sponsoring Org.:
USDOE

Primary Award/Contract Number:
AC02-05CH11231
Other Award/Contract Number:
AWD2914
Code ID:
29477
Site Accession Number:
2019-143
Research Org.:
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Country of Origin:
United States

Citation Formats

Li, Tonglin, Koziol, Quincey, and USDOE. Rootless Operations for MPI (RLO) v0.1. Computer software. https://www.osti.gov//servlets/purl/1568055. USDOE. 15 Aug. 2019. Web. doi:10.11578/dc.20191001.1.
Li, Tonglin, Koziol, Quincey, & USDOE. (2019, August 15). Rootless Operations for MPI (RLO) v0.1 [Computer software]. https://www.osti.gov//servlets/purl/1568055. doi:10.11578/dc.20191001.1.
Li, Tonglin, Koziol, Quincey, and USDOE. Rootless Operations for MPI (RLO) v0.1. Computer software. August 15, 2019. https://www.osti.gov//servlets/purl/1568055. doi:10.11578/dc.20191001.1.
@misc{osti_1568055,
title = {Rootless Operations for MPI (RLO) v0.1},
author = {Li, Tonglin and Koziol, Quincey and USDOE},
abstractNote = {Flexible communication routines that enable high-performance computing (HPC) applications to operate with less synchronization are a great benefit to application algorithm creation and developer productivity. Further, communication operations that reduce synchronization in HPC applications while continuing to scale well are critical to application performance in the exascale era. We have implemented mechanisms that allow an application to scalably execute bcast/iAllReduce from any MPI rank without specifying the origin of the message in advance, demonstrating application development flexibility, highly scalable operation, and reduced application synchronization.},
url = {https://www.osti.gov//servlets/purl/1568055},
doi = {10.11578/dc.20191001.1},
year = {2019},
month = {8},
note =
}

Software:

Save / Share: