Cluster Controller at p55cc1

Titan reboot scheduled for 23.1 already done.

Partitions maxtime restored to one week so the cluster is ready to use.

Pleione reboot scheduled for 23.1 already done.

Partitions maxtime restored to one week so the cluster is ready to use.

Due to new kernel both clusters need reboot.

All partitions' max runtime is set to 1 hour until the reboot. So, don't panic if (when) your (new) job is put on hold until the reboot and max runtime restored.

Titan is now configured according to FGCI rules. The two most significant changes are:

- HyperThreading is disabled
- 20% of resources are reserved for Grid usage

NOTE: The number of logical cores is the same as number of physical cores for all compute nodes:

ti1 :        48 cores (four 12 core CPUs)
ti2 - ti9 :  24 cores (two 12 core CPUs per node)

Pleione is fully operational now (8 new compute nodes added) and first (and last) test run completed successfully using 768 threads for one week.

System config is briefly presented on page https://p55cc-redmine.utu.fi/projects/user-s-page/wiki/Partition_Info

If you have an account in Titan you can request an account for Pleione by sending mail to fgi-admins@lists.utu.fi, your account info will be copied from Titan.

If you don't have an account in Titan request Pleione account according to this guide: https://p55cc-redmine.utu.fi/projects/user-s-page/wiki/Request_User_ID

The posting list fgi-users@lists.utu.fi has been added to receive email notification concerning NEWS of Cluster Project management system.

If you received the notification you are listed on that list and hence a cluster user. :)

Pleione has been reinstalled according to the new FGCI approach.

The whole cluster is not yet available for testing, as the new 8 compute nodes are missing.

Only test accounts have been created so far, and the cluster will be released to local usage during the following weeks.

The broken NFS disk has been replaced. Send mail to fgi-admins@lists.utu.fi in case you encounter any problems (loss of data, performance problems, etc.)

STR01/1 disk broken. Should not affect users but you never know.

Disk has been replaced and node is operational again.