29/10/2020 - CCB cluster re-open for use
The CentOS 7 cluster is now our main system:
- Two login nodes are available -
cbrglogin2.molbiol.ox.ac.uk- please use either.
- All but one of our compute nodes have been migrated to the CentOS 7 cluster.
- There is a 10-day limit on disconnected graphical logins; active graphical logins are not affected. See https://www.imm.ox.ac.uk/research/units-and-centres/mrc-wimm-centre-for-computational-biology/ccb-account for details on how to log in.
The old CentOS 6 cluster must only be used for work which cannot run on the CentOS 7 cluster:
- To maintain an equivalent level of security with our CentOS 7 cluster, you must be connected to the main University network (including Eduroam at Oxford) or an official Oxford VPN to log into deva or klyn. It is no longer possible to log in to our clusters from the open internet.
- There is only a single worker node for the CentOS 6 cluster.
- Batch jobs may not use more than 4 CPUs.
23/09/2020 - CCB cluster shutdown, week commencing 26th October
We’d like to give you advance notice of a full shutdown of the CCB bioinformatics cluster on the week of the 26th October.
Due to the COVID lockdown, we’ve postponed some essential maintenance for many months to give you maximum access to our resources during lab closures. We now need to take this work forward, which will require a full shutdown of the cluster and the login nodes.
We anticipate that the shutdown will last at least 3 days, during which time no jobs will be able to run and it will not be possible to log in. We’ll send out more specific details nearer the time.
14/09/2020 - Deva: service partially restored
The issue with the server "deva" has been traced and it is now possible to log in again. However, the root cause has been traced to a loss of connectivity to the file server which provides access to the legacy storage "/hts/data6/", which is used by a small number of older user accounts. Until the problem can be resolved, this partition is not available.
02/09/2020 - Compute Cluster CentOS 7 upgrade
We are in the process of upgrading the operating system on the CCB Compute Cluster to CentOS 7. It would be really helpful if you could find the time to logon and test the new system so we can get early feedback before we completely switch over in October/November. The advantages of the upgraded system are:
- Access to the new GPU systems
- Many package updates
- Many bioinformatics packages recompiled (use "module avail" for a list)
- Simplified Python and R module loading
- Change from NX / NoMachine to RDP for graphical login which is much more robust and easier to set up
- Change from SGE to Slurm for the batch system
03/08/2020 - Disk Storage Issues
Due to increased computational work during lockdown we are in a position where available disk storage capacity on the CCB cluster is becoming limited. We have the new hardware, which will provide the additional disk space, but the installation has been delayed due to COVID-19 access. We are now working hard to bring the storage online as soon as possible.
However, in the meantime we would appreciate it if you all could review your files on the /t1-data/ area and delete any unwanted data to make sure that we don't run out of space before we get the new hardware up and running. Also, it would be very helpful if you could keep us informed of any large new data sets (anything over 1TB) that you wish to upload to the CCB cluster by emailing email@example.com
In order to fund the new and existing storage, we are planning to bring in storage charges from April 2021. This is another good reason to try and reduce the amount stored on the system where possible. Details of pricing will follow in the autumn.