- 11 Apr, 2014 4 commits
-
-
Martins Innus authored
-
Franco Broi authored
-
David Bigagli authored
-
Morris Jette authored
Add support for allocation of GRES by model type for heterogenous systems (e.g. request a Kepler GPU, a Tesla GPU, or a GPU of any type).
-
- 10 Apr, 2014 4 commits
-
-
David Bigagli authored
not been processed by the scheduler yet.
-
Danny Auble authored
-
Morris Jette authored
Modify srun to report an exit code of zero rather than nine if some tasks exit with a return code of zero and others are killed with SIGKILL. Only an exit code of zero did this.
-
Morris Jette authored
Cache job information and drop out of loop after finding the desired array task. This improves performance a fair bit. bug 684
-
- 09 Apr, 2014 3 commits
-
-
David Bigagli authored
-
Morris Jette authored
Rather than immediately invoking an execution of the scheduling logic on every event type that can enable the execution of a new job, queue its execution. This permits faster execution of some operations, such as modifying large counts of jobs, by executing the scheduling logic less frequently, but still in a timely fashion.
-
Danny Auble authored
If you have multiple partitions the output from sinfo -o "%D %F" would have unexpected results, hardly ever correct.
-
- 08 Apr, 2014 7 commits
-
-
Morris Jette authored
-
Morris Jette authored
Fix logic bugs for SchedulerParameters option of max_rpc_cnt. Scheduling would be delayed for job arrays and backfill scheduling would be disabled unless max_rpc_cnt > 0.
-
Morris Jette authored
More gracefully handle missing batch script file. Just kill the job and do not drain the compute node.
-
Morris Jette authored
To support larger numbers of jobs when the StateSaveDirectory is on a file system that supports a limited number of files in a directory, add a subdirectory called "hash.#" based upon the last digit of the job ID.
-
Danny Auble authored
-
Danny Auble authored
on Mixed state.
-
David Bigagli authored
-
- 07 Apr, 2014 4 commits
-
-
Danny Auble authored
This changes the behavior of license_update which it's current behavior makes for doubling license counts. This is ok, because the only place it is used expects the counts to be zeroed out afterwards.
-
Morris Jette authored
-
Danny Auble authored
in it. Signed-off-by:
Danny Auble <da@schedmd.com>
-
Danny Auble authored
-
- 05 Apr, 2014 1 commit
-
-
Morris Jette authored
Disables job scheduling when there are too many pending RPCs
-
- 04 Apr, 2014 3 commits
-
-
Danny Auble authored
-
Danny Auble authored
This also reverts commit 8cff3b08 and ced2fa3f
-
Danny Auble authored
-
- 03 Apr, 2014 8 commits
-
-
Danny Auble authored
new associations were added since it was started.
-
Morris Jette authored
Permit user root to propagate resource limits higher than the hard limit slurmd has on that compute node has (i.e. raise both current and maximum limits). bug 674674674674674674
-
Morris Jette authored
Added SchedulerParameters options of bf_yield_interval and bf_yield_sleep to control how frequently and for how long the backfill scheduler will relinquish its locks.
-
Morris Jette authored
if an job step's network value is set by poe, either by directly executing poe or srun launching poe, that value was not being propagated to the job step creation RPC and the network was not being set up for the proper protocol (e.g. mpi, lapi, pami, etc.). The previous logic would only work if the srun execute line explicitly set the protocol using the --network option.
-
David Bigagli authored
-
Morris Jette authored
Permit multiple batch job submissions to be made for each run of the scheduler logic if the job submissions occur at the nearly same time. bug 616
-
Morris Jette authored
if an job step's network value is set by poe, either by directly executing poe or srun launching poe, that value was not being propagated to the job step creation RPC and the network was not being set up for the proper protocol (e.g. mpi, lapi, pami, etc.). The previous logic would only work if the srun execute line explicitly set the protocol using the --network option.
-
Morris Jette authored
Permit multiple batch job submissions to be made for each run of the scheduler logic if the job submissions occur at the nearly same time. bug 616
-
- 02 Apr, 2014 5 commits
-
-
David Bigagli authored
-
Morris Jette authored
if an job step's network value is set by poe, either by directly executing poe or srun launching poe, that value was not being propagated to the job step creation RPC and the network was not being set up for the proper protocol (e.g. mpi, lapi, pami, etc.). The previous logic would only work if the srun execute line explicitly set the protocol using the --network option.
-
Morris Jette authored
In select plugins, stop triggering extra logging based upon the debug flag CPU_Bind and use SelectType instead.
-
Morris Jette authored
-
Morris Jette authored
-
- 31 Mar, 2014 1 commit
-
-
Morris Jette authored
Address the following issues: 1A. Notes when the -V option is given and 1B. if -v option without -V then include sbatch option of "--export=none" 2A. Does not try to export environment variables by explicitly setting them in the user's environment before invoking sbatch, but instead 2B. Pass specified env vars using the sbatch --export option 3A. Recognize when qsub -v option given with key name, but no value 3B. Find the appropriate value for the specified key name and export that pair. 4. Update documentation
-