Last month we took delivery of four new Dell C6145 servers with 64 cores and 128GB of RAM each. This effectively doubles our 600 series from 256 to 512 cores. Our dashboards have undergone a facelift over the last month
…Brief HPC Headnode Outage
This morning we expierenced a brief headnode outage. Two of the physical VM nodes lost sight of a specific data store on which the hex headnode resided. The nodes were restored within 15 minutes and the headnode brought back into…
Data mining and dashboard
The current state of the dashboard is not an error. We realize that node 603 is now sitting below the other 600 series servers, however this is part of our code testing for the new dashboard which is going to…
Phase transition
Those of you who pay attention to the hex dashboard may have noticed the appearance of several new series, which are frustratingly inaccessible. ICTS engineers have been implementing a UCT cloud over the last few months and one of the…
…Hex Headnode – Downtime
Unfortunately the Hex headnode had a bit of a wobble and needed a restart. There was a brief 5 minute interruption to restart the server. The server is now available. Apologies for the downtime.…
New applications and queue limits
We've installed FastQC 0.10.1 and FastX-toolkit 0.13.2 on the hex cluster.
We've also limited the maximum number of jobs that individual users can queue at once on hex's UCTlong queue:
set queue UCTlong max_user_queuable = 50
This prohibits users
…CHPC conference and cloud training
Last week the ICTS HPC team attended the CHPC 2013 National Meeting at the CTICC. This well run and enjoyable event encouraged networking between administrators, users and vendors with plenty of very interesting talks and discussions. The CHPC's support of
…Cluster upgrades
Both clusters are back up again after the upgrade, we'd like to thank our users for their cooperation and patience over the weekend. Apart from numerous software patches we also transfered various data volumes to the new Netapp as our
…Enable FTP service for Galaxy – Web-based platform for Data intensive biomedical research
The Galaxy instance we have running at the University of Cape Town operates on the Novell SLES 11 platform. The SMT repositories do not have ProFTPD, which is the recommended FTP service for Galaxy. Also, there is a lack of …