HTCondor

license

Apache-2.0(OSI)

source

public(htcondor/htcondor is the product you run)

core-gated

ungated(a 5,213-entry recursive tree carries no path matching enterprise, ee, commercial or proprietary)

Apache-2.0, published by an academic centre with no paid edition. The 5,213-entry recursive tree carries no enterprise, ee or commercial path.

• https://github.com/htcondor/htcondor recorded 2026-09-16

  Repository page for htcondor/htcondor: Apache-2.0 license, public and unarchived, described as the HTCondor source repository. Establishes the license and that the source is public.

• https://api.github.com/repos/htcondor/htcondor/git/trees/main?recursive=1 recorded 2026-09-16

  Full untruncated recursive tree of the default branch, 5,213 entries. Licenses are the root LICENSE and vendored dependencies. No path matches enterprise, ee, commercial or proprietary. A tree lists paths; it is cited for that and for nothing about the vendor's offerings.

• https://htcondor.org/ recorded 2026-09-16

  HTCondor's own project site, run by the Center for High Throughput Computing at the University of Wisconsin-Madison. An academic project with no commercial edition positioned to withhold functionality from the published source.

The PyPI download API reports 28,301 downloads of htcondor in the trailing 30 days, inside the 10K-100K band for level 2. The package is the Python bindings rather than the scheduler daemons, and it is still the band: it covers this product and no other, so the measurement belongs to it, and preferring a star count on the grounds that bindings users are a subset of HTCondor users is a minority-share argument rather than a different-population one.

• https://pypistats.org/api/packages/htcondor/recent recorded 2026-09-16

  Trailing-window download counts for the declared PyPI artifact: 28,301 downloads in the trailing month.

Admitted on the accelerator limb: request_GPUs and require_gpus are HTCondor's own submit language, and its documentation describes 'matching jobs that need GPUs to machines that have them'. Band 5 on the parallel universe, where machine_count is 'the number of machines to be dedicated toward running the job' - a multi-machine set held for one job by HTCondor's own dedicated scheduler. Drafted at 4 from the single-job GPU submission page alone, which is silent on the parallel universe; the silence was read as a ceiling and it was not one.

• https://htcondor.readthedocs.io/en/latest/users-manual/submitting-a-job.html recorded 2026-09-16

  HTCondor's submit documentation on GPUs: 'matching jobs that need GPUs to machines that have them', with request_GPUs = <n> to declare the count and require_gpus, gpus_minimum_capability and gpus_minimum_memory to qualify which GPU. Establishes that the accelerator request is expressed in HTCondor's own schema and consumed by its matchmaking.

• https://htcondor.readthedocs.io/en/24.0/man-pages/condor_submit.html recorded 2026-09-16

  HTCondor's condor_submit manual page: 'machine_count = <max> For the parallel universe, a single value (max) is required. It is neither a maximum or minimum, but the number of machines to be dedicated toward running the job.' Establishes a multi-machine allocation held for one job.