Online Meetings and Collaboration
Meetings
What should you use for online meetings? The University supports and recommends:
Research Cluster
The Research cluster is used to support postgraduate research students and research staff. It has now been merged into the Informatics Compute Facility (ICF).
The ICF has 344 GPUs including: 48x NVIDIA RTX 2080 Ti 11GB in 6 servers (damnii07-12), 28x A40 48GB in 7 nodes (crannog01-07), 32x L40 48GB across 8 servers (scotia01-08), 8x H200 141GB partitioned into smaller virtual GPUs (saxa), and 8x H200 141GB in 1 node (herman). Please note that the configuration may change. Check the MOTD and the blog for details.
ILCC Cluster
The ILCC cluster was for members of ILCC and for CDT in NLP students.
This cluster has now been merged into the Informatics Compute Facility (ICF). The ILCC nodes are available through the slurm partitions ILCC-Standard and ILCC-CDT.
If you used the ILCC cluster prior to the merge, your previous ilcc-cluster home directory is accessible at /disk/nfs/ostrom/$USER from compute and head nodes. Going forwards you have access to /home/$USER which is on the cluster-wide lustre filesystem shared between all the nodes.
GPU cluster tips
It's easy to use a cluster the wrong way, so here are some tips on how to do it right.
See also the user-provided documentation:
Cluster architecture
To learn more, start at the GPU and Cluster Computing page.
Thanks to Antreas Antoniou for the diagram.
The teaching cluster
The Teaching cluster is used to support teaching courses, especially MSc courses and the Machine Learning Practical. It has now been merged into the Informatics Compute Facility (ICF) as the Teaching partition.
The Teaching partition has 106 GPUs in 14 servers. This includes 8x NVIDIA RTX A6000 48GB GPUs (landonia11), 90x NVIDIA RTX 2080 Ti 11GB (damnii07-12,landonias,opencast), and 8x H200 141GB partitioned into smaller virtual GPUs (saxa). Please note that the configuration may change. Check the MOTD and the blog for details.
The Slurm job scheduler
Some of the School's GPU compute clusters use the Slurm job scheduler.
Slurm matches computing jobs with computing resources. It tries to ensure that the resources are allocated fairly and that they are used efficiently. To ensure this it has complex prioritisation rules.
How to use Slurm
Slurm is widely used on supercomputers, so there are lots of guides which explain how to use it:
GPGPU computing
For a list of GPU facilities see the GPU and cluster computing page.
What is GPGPU computing?
GPGPU is general purpose computing using graphics processing units.
Charles CDT cluster
The Charles cluster is for Data Science CDT students but has been retired.
Types of hardware
This cluster has several types of computer node.
- The head node
- The head node is called
cdtcluster. To use the "charles" GPU nodes, firstssh cdtclusterthen use Slurm.Here's how to use a cluster without breaking it: