The potential of supercomputers for MD simulation runs is explored for the CYBER 205 vectorcomputer in the treatment of molecular liquids with site-site interaction potentials. Careful vectorization and optimization of remaining scalar codes is shown to reduce the CPU time to 7% as compared to the o
Vectorization of molecular dynamics Fortran programs using the cyber 205 vector processing computer
โ Scribed by R. Vogelsang; M. Schoen; C. Hoheisel
- Publisher
- Elsevier Science
- Year
- 1983
- Tongue
- English
- Weight
- 496 KB
- Volume
- 30
- Category
- Article
- ISSN
- 0010-4655
No coin nor oath required. For personal study only.
โฆ Synopsis
A concept of vectorization of molecular dynamics Fortran programs for the use of the Cyber 205 machine is presented. It is shown that for calculations with larger particle systems the program runs faster on the 205 than on the Cray-i by about a factor of two. Against conventional computers like the Cyber 175 an acceleration by a factor 10-15 is expected. A bit control vector is used instead of a neighbour list, which in principal provides calculations up to 6912 particles for the memory capacity of the Cyber 205. However, because the application of the bit vector requires computation times which grow proportional to N 2, the CPU time for particle numbers of more than 2048 becomes prohibitively large.
๐ SIMILAR VOLUMES
We present a Monte Carlo programme version written in Vector-FORTRAN 200 which allows a fast computation of thermodynamic properties of dense model fluids on the CYBER 205 vector processing computer. A comparison of the execution speed of this programme, a scalar version and a vectorized molecular
Algorithms are presented which enable molecular dynamics simulations of liquids to be performed most efficiently on the Cray-i vector processing computer. For 256 atoms interacting via the Lennard-Jones potential the CPU time per step of the simulation is 22 ms.
The set of coupled linear second-order differential equations which has to be solved for the quantum-mechanical description of inelastic scattering of atomic and nuclear particles can be rewritten as an equivalent set of coupled integral equations. This set can be solved iteratively. A concept of ve