Cluster Computing

From Solowiki
Revision as of 10:57, 26 October 2020 by Kerwin (talk | contribs) (Created page with "== Introduction == Cluster computing allows researchers to opportunity to perform computationally intense operations (e.g. fMRI analysis) on dedicated 'high performance comput...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

Introduction

Cluster computing allows researchers to opportunity to perform computationally intense operations (e.g. fMRI analysis) on dedicated 'high performance computing' HPC clusters. These clusters often have many available CPU's, GPU's en RAM storage, allowing for much faster (parallel) computing than is possible on personal / work computers.

Currently FSW users have three HPC options available:

  • Shark OLD: This HPC is maintained by the LUMC and is considered end-of-life, current users are encouraged to move to Shark SLURM
  • Shark SLURM: This is the LUMC's newer version of Shark, which is currently fully operational. FSW users can use this cluster until the new Alice cluster is fully functional
  • Alice: This is the new university-wide HPC cluster. The goal is for FSW users to exclusively use this cluster in the future. However, currently, the hardware and software needed for typical FSW use-cases (e.g. fMRI analysis) is not yet available. Users who need less specific software and have relatively low storage requirements (e.g. R, Matlab), may consider switching already.