S. Plimpton and G. Heffelfinger, Proc. of SHPCC '92, IEEE Computer Society (1992), p. 246. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
A High Performance Communications and Memory.. - Beazley, Lomdahl, .. (1994) (3 citations) (Correct)
.... most MD simulations have been limited to small simulations involving less than a million atoms[2, 3, 10, 11] However, the development of massively parallel supercomputers has generated considerable interest in developing fast parallel algorithms for performing multi million atom MD simulations [2, 3, 4, 5, 6, 7, 8]. On state of the art MPP systems, simulations with more than 100 million (10 8 ) atoms can now be performed[9] Simulations of this size will be crucial in performing realistic experiments in materials science where it will be necessary to simulate hundreds of millions or even billions of atoms ....
....communications scheme sends data in small packets using a relatively large number of message passing calls. Another approach that is commonly used is to gather all of the particle data and communicate it in several large messages. This can be done in as few as 6 message passing calls in some cases[6, 7]. However, we feel that our approach offers several advantages. First, our scheme only requires a small buffer for storing one cell of particles. If we were to communicate particle data from all of the boundary cells at once, it would require a tremendous memory overhead. For example, if each PN ....
[Article contains additional citation context not shown here]
S. Plimpton and G. Heffelfinger, Proc. of SHPCC '92, IEEE Computer Society (1992), p. 246.
Message-Passing Multi-Cell Molecular Dynamics on the.. - Beazley, Lomdahl (1993) (21 citations) (Correct)
....[7, 8] which have the potential for multi million particle simulations. In fact, recent work by Holian and collaborators [9, 10] have demonstrated 2D simulations for 1:6 Theta 10 7 atoms on a 64K processor CM 200. Work on MIMD implementaions of large scale MD is now also starting to appear [11, 12]. In this paper we present a scalable parallel MD algorithm which allows for the simulation of at least 10 8 particles interacting via a relative short range potential. We have implemented this algorithm on the Connection Machine 5 (CM 5) from Thinking Machines Corporation (TMC) and thus ....
S. Plimpton and G. Heffelfinger, Proc. of SHPCC'92, IEEE Computer Society (1992), p. 246.
GFlops Molecular Dynamics on the Connection Machine 5 - Peter Lomdahl (1993) (14 citations) (Correct)
....by the MD method, which is typically tens or maybe hundred of nano sec. at best. Ideally one would like to use the MD method for second long simulations with at least 10 9 atoms. While this goal is still very far away, there is substantial current interest in the development of fast MD algorithms[2, 3, 4, 5, 6, 7] which allow for the simulation of at least million atom systems. We have developed an new scalable MD algorithm based on a message passing multi cell approach which allows for simulating at least 10 8 particles interacting via a relative short range potential. We have implemented the algorithm ....
....or SPARC memory operations) runs at 68.5 GFlops. for r max = 2:5oe is our run with 65 million particles on 1024 PNs. In this case, the update time is 16.55 seconds which corresponds to 250 nano sec. per particle update. To the best of our knowledge, this is the best reported time to date [6]. Using minimal parallel memory caching, we were also able to simulate 180 million particles with an update time of 55.6 seconds. 10.0 100.0 1000.0 Number of Processors 0.1 1.0 10.0 100.0 r max = 2.5s r max = 5.0s 1024000 Particles Figure 5 : MD update time vs. number of processors In Fig. 5 ....
S. Plimpton and G. Heffelfinger, Proc. of SHPCC '92, IEEE Computer Society (1992), p. 246.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC