Optimize performance of HPC Systems

If you are new to running NEST on HPC systems or trying to improve the performance or debug issues, we have provided a few guides to help you out. There are a few things to consider about the hardware and software of the HPC system you are using for improving the overall performance.

Although there will be some variation between scripts and HPC systems, in general, we recommend that

• resources are fully utilized (e.g., all available cores are used).

• one thread pinned to one core (no simultaneous multithreading)

• More than one MPI process is used (for NEST 3)

Overview of HPC systems

Get an overview of common hardware and software components in HPC systems

Overview of various hardware and software components

Example SLURM script

See a typical job script, with detailed descriptions of each line.

Example Slurm script

Threading

Learn about threading and useful OpenMP settings

Threading

MPI processes

Get some tips on MPI processes and OpenMPI

MPI process

See also

See our other guides on

• Guide to parallel computing

• Benchmarking NEST

or see our performance results:

• NEST performance benchmarks