Implemented pending job and job slot limits - defaults are 2000 jobs and 32,000 job slots - applied at group level
Increased /share group quotas from 10TB to 20TB
New GPU nodes available from new_gpu LSF queue
Nvidia A10 and A100 GPUs
Nodes are running Rocky Linux 9.2
Add
. /usr/share/Modules/init/bash
to batch scripts to activate modules
Queue tuning work is ongoing to ensure partner queues are getting appropriate resources
Work that Remains In-Progress
Core Ethernet switches to be replaced
Linux to be updated to a RHEL 9 compatible distribution
new_gpu nodes running Rocky 9.2
RHEL 9 being installed on some compute nodes for testing
Additional Flex Chassis nodes from Henry2 being relocated to Hazel
Update videos to reflect new cluster name and default shell
Significant Changes between Hazel and Henry2
Running CentOS 7.9 - Henry2 had been running 7.5
Running LSF 10.1.0.13 - Henry2 had been running 10.1.0.10 [this is actually a significant difference as IBM includes significant feature changes in the minor release levels]
Default shell is bash - Henry2 had been tcsh
Began operation with significantly fewer cores than Henry2 (which had about 10,000 cores)
Operation began Jan 9 with about 3000 cores
Added about 3800 cores from Henry2 nodes (what had been racks 2b and 2l)
Added about 1600 cores with new partner nodes
There are about 4000 Flex Chassis node cores to still be moved from Henry2
Partner queues are being implemented as the associated resources are incorporated into Hazel
Some partner queues will not return as none of the associated resources will move.. all the nodes not moving are 10 years old or older