Weronika Filinger's blog

Optimising OpenMP implementation of Tinker, a molecular dynamics modelling package

Author: Weronika Filinger
Posted: 3 Jul 2014 | 11:42

Do you use scientific codes in your research? Are the things you can do with it limited by the execution time? The code has been parallelised but does not scale well? How should you go about improving the performance? What can you do when you do not have full understanding of the code? There are some general steps that can be taken to improve the performance of parallelised codes. In this article I will describe briefly the process I have undertaken to optimise the parallel performance of a computational chemistry package, TINKER, as part of the EPCC/SSI APES project.