Journal
PARALLEL COMPUTING TECHNOLOGIES (PACT 2015)
Volume 9251, Issue -, Pages 469-473Publisher
SPRINGER-VERLAG BERLIN
DOI: 10.1007/978-3-319-21909-7_45
Keywords
-
Ask authors/readers for more resources
Development of new HPC architectures proceeds faster than the corresponding adjustment of the algorithms for such fundamental mathematical models as quantum and classical molecular dynamics. There is the need for clear guiding criteria for the computational efficiency of a particular model on a particular hardware. LINPACK benchmark alone can no longer serve this role. In this work we consider a practical metric of the time-to-solution versus the computational peak performance of a given hardware system. In this metric we compare different hardware for the CP2K and LAMMPS software packages widely used for atomistic modeling. The metric considered can serve as a universal unambiguous scale that ranges different types of supercomputers.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available