Colossus in TSD: extended maintenance - FINISHED

Dear TSD-users,

the maintenance of Colossus is finished and the cluster is ready to start the queuing jobs. We took a little longer then expected and we apologise for that. The new solution is in place and all the jobs shall be submitted from /cluster/projects/pXXX and all input data shall be in /cluster/projects/pXXX. Please notice that the /tsd/pxx/data/colossus and /tsd/pxx/home partitions are no longer visible from the Colossus node. We are aware of the fact that these changes might imply a change in your jobs scripts and your workflows but it was absolutely necessary to remove a weakness in the TSD infrastructure. In addiction,  the new set up is  faster and has many positive aspects from the end-user point of view. You will soon realise it!
Please read more on the Colossus manual:
http://www.uio.no/tjenester/it/forskning/sensitiv/hjelp/hpc-usage/index.html

Please get in contact with us (tsd-drift@usit.uio.no)  if you doubt anything.

Enjoy Colossus: it is getting faster, bigger and stronger…

Regards,
Francesca

Published Apr. 12, 2016 11:12 AM