SciPost Submission Page
RUMD: A general purpose molecular dynamics package optimized to utilize GPU hardware down to a few thousand particles
by Nicholas P. Bailey, Trond S. Ingebrigtsen, Jesper Schmidt Hansen, Arno A. Veldhorst, Lasse Bøhling, Claire A. Lemarchand, Andreas E. Olsen, Andreas K. Bacher, Lorenzo Costigliola, Ulf R. Pedersen, Heine Larsen, Jeppe C. Dyre, Thomas B. Schrøder
- Published as SciPost Phys. 3, 038 (2017)
|As Contributors:||Nicholas Bailey · Lorenzo Costigliola|
|Arxiv Link:||https://arxiv.org/abs/1506.05094v4 (pdf)|
|Date submitted:||2017-11-15 01:00|
|Submitted by:||Bailey, Nicholas|
|Submitted to:||SciPost Physics|
RUMD is a general purpose, high-performance molecular dynamics (MD) simulation package running on graphical processing units (GPU's). RUMD addresses the challenge of utilizing the many-core nature of modern GPU hardware when simulating small to medium system sizes (roughly from a few thousand up to hundred thousand particles). It has a performance that is comparable to other GPU-MD codes at large system sizes and substantially better at smaller sizes.RUMD is open-source and consists of a library written in C++ and the CUDA extension to C, an easy-to-use Python interface, and a set of tools for set-up and post-simulation data analysis. The paper describes RUMD's main features, optimizations and performance benchmarks.
Ontology / TopicsSee full Ontology or Topics database.
Published as SciPost Phys. 3, 038 (2017)
Author comments upon resubmission
List of changes
1. Removed present address for one author. Several authors are no longer in academia so their present addresses are not relevant. We decided it looked better to only include people's Roskilde University affiliation (where they did the work).
2. Email address for last author TBS added.
3. First paragraph: sentence about usefulness of GPUs for MD simulation added
4. Reference added, see last sentence of third paragraph.
5. Keyword arguments (i, j) added to the example script in Figure 1 for consistency
6. References added to "Other interactions" item on Feature list, which has also been reworded slightly.
7. Clarification of the term "compute capability" in point 3 of optimization strategy list.
8. Comment added in the code in Figure 3. Some further simplication made by ommitting some lines towards the end of the function.
9. Added figure reference to first sentence in section 6 on neighborlist (order N^2)
10. Explained the 58% in section 7 in a footnote
11 Section 8 second sentence, added word "choice"
12 Fixed an incorrect value in Table 2 (should have been 192 not 196)
13 Elaborated slightly on which Ewald method we intend to implement, and included references to other GPU implementations of it.
14 Figure 7 the word "without" in the legend of Figure 7 (b) has been replaced with "with", which is more appropriate.
15 Missing s in Lennard-Jones added as noticed by second referee
Submission & Refereeing History
You are currently on this page