Posted: Wed 24 Jul 2013, 4:04pm.
The High Performance Computing cluster will be unavailable between Monday 5 and Thursday 8 August inclusive while we carry out maintenance work.
We will be adding newly purchased nodes for Engineering & Informatics, Physics and Life Sciences and upgrading the cluster to run under Scientific Linux 6.4.
In preparation for this work, we need to replace the current apollo login node with a new server. We intend to do this work this evening, Wednesday 24 July. Current users will be able to login to the cluster via feynman.hpc.susx.ac.uk whilst this work is ongoing, so you will still have access to the cluster whilst the login node is replaced and everything should continue to work as it would on apollo. Interactive jobs that are running on apollo will be shut down at 8pm this evening.
The work has been moved forward by one week to avoid any chance of disruption to computing services during the clearing process later in August.
UPDATE: 5 Aug 2013 - 12:23pm
We have delayed this week's upgrade so the HPC cluster will be available as normal today, Monday 5 August.
The upgrade will now begin at 9.00am tomorrow, Tuesday 6 August and is due to be complete by 3.00pm on Friday 9 August.
We have decided to postpone the upgrade of the HPC cluster and so the service will be available as normal this week.