Spectrum LSF

Showing 458

Allow 'ssacct' to work without specifying a file path when the session is still running

Currently Session Scheduler's 'ssacct' command works without the -f option only when the Session is finished. LSF knows where the account file is when the job is running, so why can we not see the account output when the session is running?

3 months ago in Spectrum LSF / Administration 0 Delivered

10 VOTE

Add bkill option that will set the job status to DONE rather than EXIT

Currently all jobs terminated via bkill are set to status EXIT. With the growing usage of bioinformatics tools and applications that were traditionally run on dedicated servers being integrated into use on HPC resources there is a shifting paradig...

over 2 years ago in Spectrum LSF / Administration 4 Delivered

11 VOTE

Enable bjobs to report peak CPU efficiency

Need to be able to retrieve the peak CPU efficiency from the bjobs command at the end of the run.

almost 3 years ago in Spectrum LSF / Administration 4 Delivered

7 VOTE

bkill command enhancement

We need to use bkill command to kill a specific user's pending jobs. However, there's no such a option for bkill.

over 2 years ago in Spectrum LSF / Administration 3 Delivered

9 VOTE

The number of TCP exceeds the max number 2*200 defined in LSF source code

We're getting the following message from master lim.log.doAcceptConn: Can't maintain this big clientMap[400], Connection from 10.230.236.102:36587 dropped And I received the following explanation and advice through case number TS003933691. It look...

about 3 years ago in Spectrum LSF 3 Delivered

4 VOTE

bjobs -o all

Currently there is a large list of -o options, which are hugely beneficial. We have a large bjobs -o <long list> -json report that is frequently run for metrics and query reduction (to provide an API to get cached data) but we have to update...

over 1 year ago in Spectrum LSF / Data Management 3 Delivered

2 VOTE

Add lsb.params setting to limit the number of times a job can be redispatched upon either a pre-exec failure of job initilization failure

We currently have jobs that will bounce from host to host either based upon a pre-exec failure, or a job initialization failure. This can impact > 500 hosts and cause an increase in MBD memory use. It would be nice to have a setting such as: MA...

11 months ago in Spectrum LSF / Administration 0 Delivered

3 VOTE

Enable DataManager to work properly when LSF usermapping is in place

When jobs were submitted with -data option and an account mapping is existing between the submission and execution cluster (MC) DM is not able to retrieve the data correctly.

over 1 year ago in Spectrum LSF / Other 1 Delivered

9 VOTE

Allow Stacking/Overloading of Host Closure Events

Allow the stacking of closure events. In other words, make it permissible for LSF to track multiple closure events, and release those closure events systematically.

over 5 years ago in Spectrum LSF 1 Delivered

4 VOTE

LSF integration with latest DCGM

LSF provides integration with DCGM so that GPU usage data can be collected in lsb.acct log file. However, the latest LSF only supports DCGM v1.7.2, which goes with CUDA10.For customs like us, we used CUDA11, which needs DCGM v2.1.4, and we would l...

over 2 years ago in Spectrum LSF 2 Delivered

Please enter your email address

FILTER BY CATEGORY