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

Title: On Eliminating Synchronous Communication in Molecular Simulations to Improve Scalability

Abstract

Molecular dynamics simulation, as a complementary tool to experimentation, has become an important methodology for the understanding and design of molecular systems as it provides access to properties that are difficult, impossible or prohibitively expensive to obtain experimentally. Many of the available software packages have been parallelized to take advantage of modern massively concurrent processing resources. The challenge in achieving parallel efficiency is commonly attributed to the fact that molecular dynamics algorithms are communication intensive. This paper illustrates how an appropriately chosen data distribution and asynchronous one-sided communication approach can be used to effectively deal with the data movement within the Global Arrays/ARMCI programming model framework. A new put_notify capability is presented here, allowing the implementation of the molecular dynamics algorithm without any explicit global or local synchronization or global data reduction operations. In addition, this push-data model is shown to very effectively allow hiding data communication behind computation. Rather than data movement or explicit global reductions, the implicit synchronization of the algorithm becomes the primary challenge for scalability. Without any explicit synchronous operations, the scalability of molecular simulations is shown to depend only on the ability to evenly balance computational load.

Authors:
;
Publication Date:
Research Org.:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Org.:
USDOE
OSTI Identifier:
1130716
Report Number(s):
PNNL-SA-91254
DOE Contract Number:  
AC05-76RL01830
Resource Type:
Journal Article
Journal Name:
Computer Physics Communications, 184(12):2634-2640
Additional Journal Information:
Journal Name: Computer Physics Communications, 184(12):2634-2640
Country of Publication:
United States
Language:
English

Citation Formats

Straatsma, TP, and Chavarría-Miranda, Daniel. On Eliminating Synchronous Communication in Molecular Simulations to Improve Scalability. United States: N. p., 2013. Web. doi:10.1016/j.cpc.2013.01.009.
Straatsma, TP, & Chavarría-Miranda, Daniel. On Eliminating Synchronous Communication in Molecular Simulations to Improve Scalability. United States. https://doi.org/10.1016/j.cpc.2013.01.009
Straatsma, TP, and Chavarría-Miranda, Daniel. 2013. "On Eliminating Synchronous Communication in Molecular Simulations to Improve Scalability". United States. https://doi.org/10.1016/j.cpc.2013.01.009.
@article{osti_1130716,
title = {On Eliminating Synchronous Communication in Molecular Simulations to Improve Scalability},
author = {Straatsma, TP and Chavarría-Miranda, Daniel},
abstractNote = {Molecular dynamics simulation, as a complementary tool to experimentation, has become an important methodology for the understanding and design of molecular systems as it provides access to properties that are difficult, impossible or prohibitively expensive to obtain experimentally. Many of the available software packages have been parallelized to take advantage of modern massively concurrent processing resources. The challenge in achieving parallel efficiency is commonly attributed to the fact that molecular dynamics algorithms are communication intensive. This paper illustrates how an appropriately chosen data distribution and asynchronous one-sided communication approach can be used to effectively deal with the data movement within the Global Arrays/ARMCI programming model framework. A new put_notify capability is presented here, allowing the implementation of the molecular dynamics algorithm without any explicit global or local synchronization or global data reduction operations. In addition, this push-data model is shown to very effectively allow hiding data communication behind computation. Rather than data movement or explicit global reductions, the implicit synchronization of the algorithm becomes the primary challenge for scalability. Without any explicit synchronous operations, the scalability of molecular simulations is shown to depend only on the ability to evenly balance computational load.},
doi = {10.1016/j.cpc.2013.01.009},
url = {https://www.osti.gov/biblio/1130716}, journal = {Computer Physics Communications, 184(12):2634-2640},
number = ,
volume = ,
place = {United States},
year = {Sun Dec 01 00:00:00 EST 2013},
month = {Sun Dec 01 00:00:00 EST 2013}
}