Slurm job scheduler
Slurm (Simple Linux Utility for Resource Management), formerly known as the Simple Linux Utility for Resource Management, is a free, open-source job scheduler for Linux and Unix kernel systems, widely adopted by many supercomputers and computer clusters worldwide. Slurm uses a best-fit algorithm based on Hilbert curve scheduling or fat-tree network topology to optimize the locality of task allocation on parallel computers.
It provides three key functions:
- First, it allocates exclusive or non-exclusive resources (compute nodes) for a specified duration for users to execute jobs.
- Second, it provides a framework for launching, executing, and monitoring tasks running on nodes (typically parallel tasks, such as MPI).
- Third, it reasonably allocates resources for the job queue.