I'm glad to announce you the MetaCentrum's computing capacity was extended with new clusters:
galdor.metacentrum.cz CESNET owner, 20 nodes, 1280 CPU cores aand 80x GPU NVIDIA A40, in each node:
The cluster can be accessed via the conventional job submission through PBS batch system (@pbs-meta server) in gpu priority and short default queues.
On GPU clusters, it is possible to use Docker images from NVIDIA GPU Cloud (NGC) - the most used environment for the development of machine learning and deep learning applications, HPC applications or visualization accelerated by NVIDIA GPU cards. Deploying these applications is then a matter of copying the link to the appropriate Docker image, running it in the Docker container in Singularity. More information can be found at https://wiki.metacentrum.cz/wiki/NVidia_deep_learning_frameworks
halmir.metacentrum.cz CESNET, 31 nodes, 1984 CPU cores, in each node:
The cluster can be accessed via the conventional job submission through PBS batch system (@pbs-meta server) in short default queues. Longer queues will be added after testing.
We continuously solve problems with the compatibility of some applications with the Debian11 OS by recompiling new SW modules. If you encounter a problem with your application, try adding the debian10-compat module at the beginning of the startup script. If the problems persist, let us know at meta (at) cesnet.cz.
For complete list of available HW in MetaCentrum see http://metavo.metacentrum.cz/pbsmon2/hardware