OpenPBS

OpenPBS is a workload manager and job scheduler for high-performance computing. It optimizes how jobs are scheduled and resources are used across clusters, clouds and supercomputers, helping research and engineering sites improve system efficiency and user productivity. In May 2020 the PBS Professional open-source project was renamed OpenPBS, and the commercial product keeps the PBS Professional name, with the same underlying code.

The README highlights scalability, with fast job dispatch and low latency at very large core counts, and policy-driven scheduling that balances turnaround time and utilization to meet site goals and service levels. It also lists resiliency through automatic fail-over so jobs are not lost during failures, a flexible plugin framework with hooks for customization, and a health-check framework that monitors and mitigates faults.

The software is written in C and has been used for many years in both commercial and public-sector sites. The project is a community effort with several ways to participate, and its website is the starting point for new users. The repository metadata lists the license as Other, so review the license file.