Latest news on the DEEP-EST prototype system

This is a summary of the latest news concerning the system. For a list of known problems related to the system, please refer to this page.

System hardware

CM nodes

ESB nodes

• currently only 1 rack installed in the system (25 nodes).
• installation of the 2 remaining racks is scheduled for mid of May

DAM nodes

• all nodes still run in app-direct mode for the DCPMM (see #2366)

Network Federation Gateways

• two gateway nodes for IB ↔ Extoll bridging between CM/ESB and DAM are available for the users

• for an example on how to use the gateway nodes and for further information, please refer to the batchsystem wiki page.

Global resources

File Systems

• The old shared file system /sdv-work (from previous DEEP projects) currently does not work. Metadata are still available (files and directories can be seen), but data is not accessible. Please contact deepest-sup@… if you need to recover data from the file system.

• Still ongoing investigations on the performance of the shared file system /work (BeeGFS).

• It is now possible to access the $ARCHIVE file system from the deepv login node under /arch. For more information about $ARCHIVE, please refer to the Filesystems page.

System software

SW updates

• new SLURM version 19.5.5 with workflow support has been rolled out
  • information on how to use workflows have been added here.
  • please, report any problems found

• new ParaStationMPI version (5.4.5-2) is installed and used as default now
  • Pending: Install software packages that depend on psmpi also with the new version

User management

QoS

• the quality of service groups (QoS) within Slurm (determining the jobs priority) are currently being re-organized with regards to the accounting introduced with the EAP

• please, report any issues in starting jobs or in specifying an account for your jobs using the -A option

BeeGFS Quotas

• it's currently under consideration to implement a quota for the BeeGFS file system (mounted to /work)