Description
The training focuses on different ways to allocate resources on supercomputers and parallelisation schemes used to speed up calculations.
Modern supercomputers are massive parallel machines with up to hundreds of cores on a single node and even thousands of nodes connected via fast network interface. To utilise those machines fully, one needs to know about parallelisation techniques and queuing system options of resource allocation. This training will show different ways that Slurm allocates resources and how to apply them to different computations and workloads.
Date
6 November 2026, 10:00 - 13:00 (online)
Agenda
- Introduction - serial vs parallel,
- processes, htop and interactive access to a job:
- processes vs threads,
- OpenMP vs MPI,
- hybrid execution,
- mpirun vs srun;
- binaries - OpenMP, MPI, both, neither? - how to check app's libraries,
- limits of parallelism - matching execution to algorithm,
- GROMACS - one application, countless execution methods,
- accelerating GROMACS with a GPU card,
- Q&A (15 min).
