Jump to letter: [
CGHIJLMNOPVX
]
condor - HTCondor: High Throughput Computing
- Description:
HTCondor is a specialized workload management system for
compute-intensive jobs. Like other full-featured batch systems, HTCondor
provides a job queuing mechanism, scheduling policy, priority scheme,
resource monitoring, and resource management. Users submit their
serial or parallel jobs to HTCondor, HTCondor places them into a queue,
chooses when and where to run the jobs based upon a policy, carefully
monitors their progress, and ultimately informs the user upon
completion.
Packages
| condor-24.0.2-1.el8.aarch64
[9.0 MiB] |
Changelog
by Tim Theisen (2024-11-26):
- Add STARTER_ALWAYS_HOLD_ON_OOM to minimize confusion about memory usage
- Fix bug that caused condor_ssh_to_job sftp and scp modes to fail
- Fix KeyboardIdle attribute in dynamic slots that could prevent job start
- No longer signals the OAuth credmon when there is no work to do
- Fix rare condor_schedd crash when a $$() macro could not be expanded
- By default, put Docker jobs on hold when CPU architecture doesn't match
|
| condor-24.0.1-1.el8.aarch64
[9.0 MiB] |
Changelog
by Tim Theisen (2024-10-31):
- Improved tracking and enforcement of disk usage by using LVM
- Enhancements to the htcondor CLI tool
- cgroup v2 support for tracking and enforcement of CPU and memory usage
- Leverage cgroups to hide GPUs not allocated to the job
- DAGMan can now produce job credentials when using direct submit
- New submit commands to aid in matching specific GPU requirements
- New implementation of the Python bindings, htcondor2 and classad2
- Improved default security configuration
- Significant reduction in memory and CPU usage on the Central Manager
- Support for GPUs using AMD's HIP 6 library
- Fix bugs when -divide or -repeat was used in GPU detection
- Proper error message and hold when Docker emits multi-line error message
- Fix issue where an unresponsive libvirtd blocked an EP from starting up
|