[Jlab-scicomp-briefs] JLab Data Center chilled water supply repair work

[Bryan Hess]

Jefferson Lab facilities management discovered a leak in the chilled water supply pipe for data center cooling. Investigating and repairing the leak requires that the data center operate on a rental chiller, which is now installed and connected outside CEBAF Center.


The JLab Facilities Management & Logistics team will switch over from the data center's current chilled water source to this rental unit tomorrow, 8/2/22, at 8AM. The work will begin with a 48-hour reliability test of the temporary chiller before the main chilled water lines are taken out of service.  After the 48-hour test, digging and repair work will begin, and the temporary chiller will be the sole source of cooling for the farm and HPC portion of the data center.


The highest outage risk in case of a cooling failure is to compute nodes and the tape library. The core computing functions, including business systems, networking, and file servers have a second, fully redundant cooling system that does not depend on the chilled water supply to remain operational.


Although no outage is anticipated, there is an increased risk of a cooling failure affecting the compute farm and LQCD clusters. In the event of a cooling outage, clusters will be shut down until cooling is restored and users will be notified. There is no action needed on the part of scientific computing users. The temporary chiller will remain in place for the duration of the repair, which could take a few weeks to complete.



Please direct any questions to Bryan Hess, bhess at jlab.org<mailto:bhess at jlab.org>.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/jlab-scicomp-briefs/attachments/20220801/7daa66c1/attachment.html>