Spectrum LSF

Showing 484

Modify cgroup memory limit of running job

With "bsub -M ..." LSF can use cgroups to limit the amount of memory a job can use. Unfortunately, bmod cannot change this for running, despite the cgroups do support modifying this.

almost 3 years ago in Spectrum LSF / Scheduling 2 Delivered

4 VOTE

Provide capability to view jobs submission environment

In prior releases, if an administrator needed to investigate a submission script, they could goto the LSF info directory and review the job script. However, this is not longer possible using the in memory algorithm. This functionality needs to be ...

over 5 years ago in Spectrum LSF 1 Delivered

1 VOTE

hostcache auto-delete function in the mbatchd restart

Dynamic hosts remain in the cluster unless you intentionally remove them.And the manual related to it is as follows.https://www.ibm.com/docs/en/spectrum-lsf/10.1.0?topic=cluster-remove-dynamic-hostsSuppose we delete the hostcache file, not modify ...

over 1 year ago in Spectrum LSF / Administration 2 Delivered

2 VOTE

LSF/GPFS monitoring and job submission

Nvidia would like a documented/supported methodology to ensure lsf and scale/gpfs nsd nodes are in healthy state prior to job submission.

almost 3 years ago in Spectrum LSF 2 Delivered

1 VOTE

Have LSF Handle ZOMBI job searching using either BTREE or HASH ALGO's

Large number of ZOMBI jobs can impact very large clusters operations resulting in high backlogs and additional lost job status. Improving the ZOMBI search algorithm from Linear Search to BTREE or HASH would reduce this impact.

over 1 year ago in Spectrum LSF / Administration 1 Delivered

1 VOTE

Make the TCP Backlog for the MBATCH and MBATCHD Query Port Configurable for Large Clusters

We have been monitoring the TCP backlogs for LIM, MBD, and the Query port for some time, and there are periods where the MBD query port becomes overloaded resulting in a higher than expected count of UNKNOWN and sometimes eventually ZOMBI jobs, an...

over 1 year ago in Spectrum LSF / Administration 1 Delivered

4 VOTE

allow lsb.stream files to use data from already existing lsb.acct

If the lsb.stream file not generated or has been removed, LSF Explorer can't load the job data into ES. Currently LSF Explorer doesn't have the functionality to load the data from lsb.acct files, but it is possible to re-generate lsb.stream file b...

almost 6 years ago in Spectrum LSF 2 Delivered

3 VOTE

Add panel to modify Fairshare configuration in LSF Application Center (PAC)

Spectrum LSF Application Center provided the capability to graphically view the queue fairshare usage in a specific period of time. The users are working as groups and the fairshare policies are defined at beginning of each week by an agreement be...

over 4 years ago in Spectrum LSF 1 Delivered

3 VOTE

Job reached run limit & get kill , user need notification 48 hr /24 hr /12 hr before LSF kill job

Job reached run limit & get kill , user need notification 48 hr /24 hr /12 hr before LSF kill job can we talk on this. It seems user get notification only XX min before kill ie customer number Today in RTM we have only 1 option say select % bu...

over 4 years ago in Spectrum LSF 2 Delivered

3 VOTE

Fairshare should include gpu factors

We would like gpu factors to be considered in the fairshare structure.

over 4 years ago in Spectrum LSF 0 Delivered

Please enter your email address

FILTER BY CATEGORY