Simple Linux Utility for Resource Management

Overview Repositories Revisions Requests Users Attributes Meta

Simple Linux Utility for Resource Management

SLURM is an open source, fault-tolerant, and highly scalable cluster management and job scheduling system for Linux clusters
containing up to 65,536 nodes. Components include machine status, partition management, job management, scheduling and accounting modules.

Developed at network:cluster
Sources inherited from project openSUSE:Factory
3 derived packages
Derived Packages
network:cluster

home:dirkmueller:AL:TW

security:tls:staging
Cancel
Download package
Checkout Package
osc -A https://api.opensuse.org checkout home:adrianSuSE:PL/slurm && cd $_
Create Badge

Build Results
RPM Lint

Refresh

Source Files

Filename	Size	Changed
Remove-rpath-from-build.patch	0000002941 2.87 KB	over 4 years ago
_service	0000000133 133 Bytes	over 7 years ago
check-for-lipmix.so.MAJOR.patch	0000001425 1.39 KB	over 4 years ago
pam_slurm-Initialize-arrays-and-pass-sizes.patch	0000002549 2.49 KB	over 3 years ago
slurm-2.4.4-init.patch	0000001132 1.11 KB	over 11 years ago
slurm-20.11.0.tar.bz2	0006475238 6.18 MB	over 3 years ago
slurm-rpmlintrc	0000000761 761 Bytes	about 4 years ago
slurm.changes	0000091602 89.5 KB	over 3 years ago
slurm.spec	0000036372 35.5 KB	over 3 years ago

Revision 50 (latest revision is 105)

Dominique Leuenberger (dimstar_suse) accepted request 853268 from

Egbert Eich (eeich) over 3 years ago (revision 50)

- Update to version 20.11.0
  Slurm 20.11 includes a number of new features including:
  * Overhaul of the job step management and launch code, alongside improved
    GPU task placement support.
  * A new "Interactive Step" mode of operation for salloc.
  * A new "scrontab" command that can be used to submit and manage
    periodically repeating jobs.
  * IPv6 support.
  * Changes to the reservation logic, with new options allowing users
    to delete reservations, allowing admins to skip the next occurance of a
    repeated reservation, and allowing for a job to be submitted and eligible
    to run within multiple reservations.
  * Dynamic Future Nodes - automatically associate a dynamically
    provisioned (or "cloud") node against a NodeName definition with matching
    hardware.
  * An experimental new RPC queuing mode for slurmctld to reduce thread
    contention on heavily loaded clusters.
  * SlurmDBD integration with the Slurm REST API.
  Also check
  https://github.com/SchedMD/slurm/blob/slurm-20-11-0-1/RELEASE_NOTES (forwarded request 852039 from eeich)