dorsal/arxiv
View SchemaParallel computing and molecular dynamics of biological membranes
| Authors | G. La Penna, S. Letardi, V. Minicozzi, S. Morante, G. C. Rossi, G. Salina |
|---|---|
| Categories | |
| ArXiv ID | physics/9709024 |
| URL | https://arxiv.org/abs/physics/9709024 |
| DOI | 10.1016/S0920-5632(97)00963-8 |
Abstract
In this talk I discuss the general question of the portability of Molecular Dynamics codes for diffusive systems on parallel computers of the APE family. The intrinsic single precision arithmetics of the today available APE platforms does not seem to affect the numerical accuracy of the simulations, while the absence of integer addressing from CPU to individual nodes puts strong constraints on the possible programming strategies. Liquids can be very satisfactorily simulated using the "systolic" method. For more complex systems, like the biological ones at which we are ultimately interested in, the "domain decomposition" approach is best suited to beat the quadratic growth of the inter-molecular computational time with the number of elementary components of the system. The promising perspectives of using this strategy for extensive simulations of lipid bilayers are briefly reviewed.
{
"annotation_id": "7d497f5e-9be2-45bc-bb53-49314fcd79a5",
"date_created": "2026-03-02T18:01:21.790000Z",
"date_modified": "2026-03-02T18:01:21.790000Z",
"file_hash": "137f7070affcfd2bc018aa5f9a144877601a4a94c68edb69f94d19821fe23ca2",
"private": false,
"record": {
"abstract": "In this talk I discuss the general question of the portability of Molecular\nDynamics codes for diffusive systems on parallel computers of the APE family.\nThe intrinsic single precision arithmetics of the today available APE platforms\ndoes not seem to affect the numerical accuracy of the simulations, while the\nabsence of integer addressing from CPU to individual nodes puts strong\nconstraints on the possible programming strategies. Liquids can be very\nsatisfactorily simulated using the \"systolic\" method. For more complex systems,\nlike the biological ones at which we are ultimately interested in, the \"domain\ndecomposition\" approach is best suited to beat the quadratic growth of the\ninter-molecular computational time with the number of elementary components of\nthe system. The promising perspectives of using this strategy for extensive\nsimulations of lipid bilayers are briefly reviewed.",
"arxiv_id": "physics/9709024",
"authors": [
"G. La Penna",
"S. Letardi",
"V. Minicozzi",
"S. Morante",
"G. C. Rossi",
"G. Salina"
],
"categories": [
"physics.bio-ph",
"cond-mat.soft",
"hep-lat",
"q-bio"
],
"doi": "10.1016/S0920-5632(97)00963-8",
"title": "Parallel computing and molecular dynamics of biological membranes",
"url": "https://arxiv.org/abs/physics/9709024"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "ca21e424-a3e9-4dd2-b1a5-48f8a55b40ac",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}