- 05 Jan, 2016 1 commit
-
-
Morris Jette authored
Don't decrease a pool's "used_space" below the amount we believe has been allocated in case there is a buffer allocation in progress while the _load_state() function is running. Initialize a mutex Move loading of state_save_location earlier
-
- 04 Jan, 2016 1 commit
-
-
Morris Jette authored
-
- 31 Dec, 2015 1 commit
-
-
Morris Jette authored
If burst buffer space is reserved through an advanced reservation then reduce the space reserved by the amount of space already allocated to jobs running in that reservation.
-
- 30 Dec, 2015 4 commits
-
-
Morris Jette authored
-
Morris Jette authored
Previous logic had a window between a job's allocation of burst buffer resources and those resources being recorded in use. During that interval, those same resources could be allocated (again) to another job. This change eliminates the race condition.
-
Morris Jette authored
-
Morris Jette authored
-
- 29 Dec, 2015 6 commits
-
-
Morris Jette authored
Make advanced reservation logic work for altternate pool names. If no pool specified, use default pool
-
Morris Jette authored
Burst buffer advanced reservation units treated as bytes (per documentation) rather than GB.
-
Morris Jette authored
Rename avail_space to total_space and correct value displayed by scontrol show burst
-
Morris Jette authored
Support per-pool granularity logic
-
Morris Jette authored
Rename "gres" to "pool" in some data structures Remove Plugin from default sview burst buffer page Change burst_buffer.conf field from "Gres" to "AltPool" Rename avail_cnt and used_cnt to avail_space and used_space respectively No other changes to logic
-
Morris Jette authored
-
- 28 Dec, 2015 1 commit
-
-
Morris Jette authored
This is so we can more easily change the data structures for burst_buffer/cray without worrying about the generic plugin, which is a "throw-away" at this point.
-
- 22 Dec, 2015 2 commits
-
-
Morris Jette authored
Major refactoring of the burst_buffer/cray check for available space (before allocating a new buffer) including preemption logic. These changes are needed to support multiple pools.
-
Morris Jette authored
-
- 21 Dec, 2015 2 commits
-
-
Morris Jette authored
Interactive pools support state save/restore fix
-
Morris Jette authored
-
- 17 Dec, 2015 1 commit
-
-
Morris Jette authored
Remove vestigial common functions for persistent buffer management Get utilization calculation by pool working
-
- 16 Dec, 2015 12 commits
-
-
Morris Jette authored
Add granularity info to alternate pools Also change the scontrol show burst buffer format for bb/GRES: " Gres[%d] Name=%s AvailCount=%s UsedCount=%s" to " AltPoolName[%d]=%s Granularity=%s TotalSpace=%s UsedSpace=%s" for better consistency with default pool information Add pool name to burst buffer creation data structures Create persistent buffer in user-specified pool Add pool info to allocated buffer table Fix buffer state load logic, was creating duplicate pool/gres records bug 2190
-
Morris Jette authored
When extra pools were reported by Cray API, it was not multiplying the buffer count by granularity to get the total size, but was just getting the counts.
-
Morris Jette authored
-
Morris Jette authored
Move slurmctld mail handler to separate thread for improved performance. Original logic did fork/exec without separate thread and if the slurmctld memory size is huge, then the time required for fork() to complete can be significant. bug 2252
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
-
Morris Jette authored
Conflicts: src/slurmctld/job_scheduler.c
-
Morris Jette authored
Conflicts: src/slurmctld/job_mgr.c
-
Morris Jette authored
Avoid hard-link/copy of script/environment files for job arrays. Use the master job record file for all tasks of the job array. back port of commit 41eead64 bug 2220
-
Morris Jette authored
Refactor logic to synchronize active bach jobs and their script/environment files, reducing overhead dramatically for large numbers of active jobs. backport of commit 41eead64 bug 2220
-
- 15 Dec, 2015 9 commits
-
-
Morris Jette authored
-
Morris Jette authored
If a job can never be started due to unsatisfied job dependencies, report the full original job dependency specification rather than the dependencies remaining to be satisfied (typically NULL). bug 2197
-
Danny Auble authored
in use.
-
Brian Christiansen authored
-
Brian Christiansen authored
-
Danny Auble authored
Conflicts: src/slurmctld/proc_req.c
-
Danny Auble authored
_slurm_rpc_complete_job_allocation. This is a rewrite of 438365ec which didn't catch the job_ptr wasn't in a lock so the memory issue could still of existed. This hopefully fixes all the spots the job_ptr wasn't in the lock. Fixes bug 2146
-
Danny Auble authored
This reverts commit 438365ec.
-
Brian Christiansen authored
Bug 2261
-