Data Centre - service suspension

We've now resolved the incident and the service is fully restored to 100% capacity.

Thanks for your patience.

The service is currently degraded. GPU capacity is operating at approximately 75%, while CPU capacity remains at around 50%.

Our plan to restore the service to full capacity next week remains on track.

Thank you for your patience.

The service is currently degraded and operating at approximately 50% capacity (up from 25% yesterday).

The login node and storage arrays are working normally, and data can be accessed or retrieved as usual.

At this stage, we expect full service restoration by Monday.

We restored 5 of the 40 nodes on Monday morning and are monitoring the performance this week with a view to a full restoration by Friday.

The air conditioning unit has been restored. However, we are keeping the compute nodes off overnight and will do a phased recovery tomorrow. Please await a further update in the morning and apologies again for the disruption of service.

We are currently experiencing an air-conditioning issue that requires the shutdown of some services including HPC (DMOG). We are working to restore them as a matter of urgency and will update this page with updates. Apologies for the inconvenience.