Performance Models on QCDOC for Molecular Dynamics with Coulomb Potentials

Authors:
Y. Deng;J. Glimm;J. W. Davenport;X. Cai;E. Santos
Affiliations:
DEPARTMENT OF APPLIED MATHEMATICS, STONY BROOK UNIVERSITY, STONY BROOK, NY 11794, USA AND CENTER FOR DATA INTENSIVE COMPUTING, BROOKHAVEN NATIONAL LABORATORY, UPTON, NY 11973, USA;DEPARTMENT OF APPLIED MATHEMATICS, STONY BROOK UNIVERSITY, STONY BROOK, NY 11794, USA AND CENTER FOR DATA INTENSIVE COMPUTING, BROOKHAVEN NATIONAL LABORATORY, UPTON, NY 11973, USA;CENTER FOR DATA INTENSIVE COMPUTING, BROOKHAVEN NATIONAL LABORATORY, UPTON, NY 11973, USA;UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA, HEFEI 230026, CHINA;DEPARTMENT OF COMPUTER SCIENCE, VIRGINIA TECH., BLACKSBURG, VA 24061, USA
Venue:
International Journal of High Performance Computing Applications
Year:
2004

Citing 7
Cited 1

Computer simulation of liquids

Computer simulation of liquids
Computer simulation using particles

Computer simulation using particles
NAMD2: greater scalability for parallel molecular dynamics

Journal of Computational Physics - Special issue on computational molecular biophysics
Understanding Molecular Simulation: From Algorithms to Applications

Understanding Molecular Simulation: From Algorithms to Applications
NAMD: biomolecular simulation on thousands of processors

Proceedings of the 2002 ACM/IEEE conference on Supercomputing
Blue Gene: a vision for protein science using a petaflop supercomputer

IBM Systems Journal - Deep computing for the life sciences
Parallel Particle Simulations of Thin-Film Deposition

International Journal of High Performance Computing Applications

Electrostatic force computation for bio-molecules on supercomputers with torus networks

Parallel Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We estimate that a novel architecture massively parallel computer, the QCDOC, can integrate molecular dynamics equations for 105 particles interacting via long-range forces (including Coulomb) for 1-10 s of simulated time using several weeks of computing time using 8000 or 10,000 processors. This number of atoms is typical for biological molecules. The two main conclusions we reach are as follows. (1) This is an increase of more than one order of magnitude in simulated time over current simulations. (2) The novel architecture, with 24 parallel channels of low latency communication per processor, allows improved long-range communication and an unusual degree of fine-scale parallelism, compared to conventional switch-based architectures. The technical heart of the paper is a detailed analysis of the computing time used in the Ewald method as a function of the required accuracy, the size of the molecular dynamics cell, and the hardware design parameters.