Skip to main content.

Computing with HTCondor™

Our goal is to develop, implement, deploy, and evaluate mechanisms and policies that support High Throughput Computing (HTC) on large collections of distributively owned computing resources. Guided by both the technological and sociological challenges of such a computing environment, the Center for High Throughput Computing at UW-Madison has been building the open source HTCondor distributed computing software (pronounced "aitch-tee-condor") and related technologies to enable scientists and engineers to increase their computing throughput.

Note: The HTCondor software was known as 'Condor' from 1988 until its name changed in 2012. If you are looking for Phoenix Software International's software development and library management system for z/VSE or z/OS, click here.

Latest News [RSS]

HTCondor-CE 4.1.0 released!

January 9, 2020

The HTCondor team is pleased to announce the release of HTCondor-CE 4.1.0. HTCondor-CE is a grid gateway that forwards resource provisioning requests to your local cluster. HTCondor-CE was originally developed by the Open Science Grid (OSG). The Center for High-Throughput Computing (CHTC) now distributes and supports HTCondor-CE for use outside of the OSG. Highlights of this release are: Added the ability to configure the environment of routed jobs; Simplified APEL configuration; Fixed the ability to specify certificate locations for SSL authentication; Refined the APEL record filter to ignore jobs that have not yet started; Fixed an issue where "condor_ce_q" required authentication; Re-enabled the ability for local users to submit jobs to the CE queue; Fixed an issue where some jobs were capped at 72 mins instead of 72 hrs; Improved BDII provider error handling. A complete list of new features and fixed bugs can be found in the version history. HTCondor-CE 4.1.0 binaries are available from our YUM Repository.
HTCondor-CE 3.4.0 released!

January 9, 2020

The HTCondor team is pleased to announce the release of HTCondor-CE 3.4.0. HTCondor-CE is a grid gateway that forwards resource provisioning requests to your local cluster. HTCondor-CE was originally developed by the Open Science Grid (OSG). The Center for High-Throughput Computing (CHTC) now distributes and supports HTCondor-CE for use outside of the OSG. Highlights of this release are: Added the ability to configure the environment of routed jobs; Simplified APEL configuration; Refined the APEL record filter to ignore jobs that have not yet started; Improved BDII provider error handling. A complete list of new features and fixed bugs can be found in the version history. HTCondor-CE 3.4.0 binaries are available from our YUM Repository.
HTCondor 8.9.5 released!

January 2, 2020

The HTCondor team is pleased to announce the release of HTCondor 8.9.5. This development series release contains new features that are under development. This release contains all of the bug fixes from the 8.8.7 stable release. Enhancements in the release include: Added a new mode that skips jobs whose outputs are newer than their inputs; Added command line tool to help debug ClassAd expressions; Added port forwarding to Docker containers; You may now change some DAGMan throttles while the DAG is running; Added support for session tokens for pre-signed S3 URLs; Improved the speed of the negotiator when custom resources are defined; Fixed interactive submission of Docker jobs; Fixed a bug where jobs would not be killed when getting an OOM notification. Further details can be found in the Development Version History and the Stable Version History. HTCondor 8.9.5 binaries and source code are available from our Downloads page.
HTCondor 8.8.7 released!

December 26, 2019

The HTCondor team is pleased to announce the release of HTCondor 8.8.7. A stable series release contains significant bug fixes. Highlights of this release are: Updated condor_annex to work with upcoming AWS Lambda function changes; Added the ability to specify the order that job routes are applied; Fixed a bug that could cause remote condor submits to fail; Fixed condor_wait to work when the job event log is on AFS; Fixed RPM packaging to be able to install condor-all on CentOS 8; Period ('.') is allowed again in DAGMan node names. More details about the fixes can be found in the Version History. HTCondor 8.8.7 binaries and source code are available from our Downloads page.
Using 50k GPUs in one HTCondor cloud-based pool

December 6, 2019

Read about how at SC19 UCSD got over 50k GPUs from the three largest Cloud providers, and integrated them in a single a HTCondor pool to run production science jobs. At peak, this cloud-based dynamic HTCondor cluster provided more than 90% of the performance of the #1 TOP100 HPC supercomputer.
HTCondor 8.9.4 released!

November 19, 2019

The HTCondor team is pleased to announce the release of HTCondor 8.9.4. This development series release contains new features that are under development. This release contains all of the bug fixes from the 8.8.6 stable release. Enhancements in the release include: Amazon S3 file transfers using pre-signed URLs; Further reductions in DAGMan memory usage; Added -idle option to condor_q to display information about idle jobs; Support for SciTokens authentication; A tool, condor_evicted_files, to examine the SPOOL of an idle job. Further details can be found in the Development Version History and the Stable Version History. HTCondor 8.9.4 binaries and source code are available from our Downloads page.

Save the dates!

HTCondor Week 2020 will be held Monday, May 18 through Thursday, May 21 at the Fluno Center on the University of Wisconsin-Madison campus. Registration will open in February. Check back here for more details.

Software

Community

Research and Development