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

Title: Global-view coefficients: a data management solution for parallel quantum Monte Carlo applications: A DATA MANAGEMENT SOLUTION FOR QMC APPLICATIONS

Journal Article · · Concurrency and Computation. Practice and Experience
DOI:https://doi.org/10.1002/cpe.3748· OSTI ID:1372691
 [1]; ORCiD logo [2];  [1];  [3];  [4];  [5];  [6];  [1]
  1. Department of Computer Science and Engineering, The Ohio State University, Columbus OH USA
  2. Mathematics and Computer Science Division, Argonne National Laboratory, Lemont IL USA
  3. Argonne Leadership Computing Facility, Argonne National Laboratory, Lemont IL USA
  4. Materials Science and Technology and Computational Chemistry and Materials Division, Oak Ridge National Laboratory, Oak Ridge TN USA
  5. Deptartment of Physics, North Carolina State University, Raleigh NC USA
  6. Deptartment of Physics, University of Illinois at Urbana-Champaign, Champaign IL USA

Quantum Monte Carlo (QMC) applications perform simulation with respect to an initial state of the quantum mechanical system, which is often captured by using a cubic B-spline basis. This representation is stored as a read-only table of coefficients and accesses to the table are generated at random as part of the Monte Carlo simulation. Current QMC applications, such as QWalk and QMCPACK, replicate this table at every process or node, which limits scalability because increasing the number of processors does not enable larger systems to be run. We present a partitioned global address space approach to transparently managing this data using Global Arrays in a manner that allows the memory of multiple nodes to be aggregated. We develop an automated data management system that significantly reduces communication overheads, enabling new capabilities for QMC codes. Experimental results with QWalk and QMCPACK demonstrate the effectiveness of the data management system.

Research Organization:
Argonne National Laboratory (ANL), Argonne, IL (United States)
Sponsoring Organization:
National Science Foundation (NSF); USDOE Office of Science (SC), Basic Energy Sciences (BES)
DOE Contract Number:
AC02-06CH11357
OSTI ID:
1372691
Journal Information:
Concurrency and Computation. Practice and Experience, Vol. 28, Issue 13; ISSN 1532-0626
Publisher:
Wiley
Country of Publication:
United States
Language:
English

References (8)

Towards Scalable Parallelism in Monte Carlo Particle Transport Codes Using Remote Memory Access journal January 2011
Advances, Applications and Performance of the Global Arrays Shared Memory Programming Toolkit journal May 2006
ScaLAPACK Users' Guide book January 1997
Titanium: a high-performance Java dialect journal September 1998
ScalaBLAST: A Scalable Implementation of BLAST for High-Performance Data-Intensive Bioinformatics Analysis journal August 2006
QWalk: A quantum Monte Carlo program for electronic structure journal May 2009
Parallel Programmability and the Chapel Language journal August 2007
ARMCI: A portable remote memory copy library for distributed array libraries and compiler run-time systems book January 1999

Similar Records

Global-View Coefficients: A Data Management Solution for Parallel Quantum Monte Carlo Applications
Conference · Tue Jan 01 00:00:00 EST 2013 · OSTI ID:1372691

Global‐view coefficients: a data management solution for parallel quantum Monte Carlo applications
Journal Article · Thu Jan 28 00:00:00 EST 2016 · Concurrency and Computation. Practice and Experience · OSTI ID:1372691

Quantum Monte Carlo Endstation for Petascale Computing
Technical Report · Wed Mar 02 00:00:00 EST 2011 · OSTI ID:1372691