Simple Linux Utility for Resource Management
Simple Linux Utility for Resource Management (SLURM) is an open source job scheduler used by many of the world's supercomputers and computer clusters. It provides three key functions. First, it allocates exclusive and/or non-exclusive access to resources (computer nodes) to users for some duration of time so they can perform work. Second, it provides a framework for starting, executing, and monitoring work (typically a parallel job such as MPI) on a set of allocated nodes. Finally, it arbitrates contention for resources by managing a queue of pending jobs.
SLURM is the batch system on many of the TOP500 supercomputers, including the second fastest one in the world IBM Sequoia, a 20-petaflop IBM BlueGene/Q with 1.6 million cores and has scheduled machines over an order of magnitude larger than this using emulation.
SLURM uses a best fit algorithm based on Hilbert curve scheduling or fat tree network topology in order to optimize locality of task assignments on parallel computers.[1]
History
SLURM began development as a collaborative effort primarily by Lawrence Livermore National Laboratory, SchedMD, Linux NetworX, Hewlett-Packard, and Groupe Bull as a Free Software resource manager. Over 100 people around the world have contributed to the project. It has since evolved into a sophisticated batch scheduler capable of satisfying the requirements of many large computer centers. SLURM is currently used on many of the largest computers in the world including Sequoia (a 20 PetaFlop IBM Bluegene/Q with 1.6 million cores at Lawrence Livermore National Laboratory), Stampede (a 2.66 PetaFlop Dell computer at the Texas Advance Computing Center), Tianhe-1A (a 2.51 PetaFlop system at the National University of Defense Technology, China), and Curie (a 2.00 PetaFlop Bull system at the CEA, France).
Structure
SLURM's design is very modular with dozens of optional plugins. In its simplest configuration, it can be installed and configured in a couple of minutes. More sophisticated configurations provide database integration for accounting, management of resource limits and workload prioritization. SLURM also works with several meta-schedulers such as Moab Cluster Suite, Maui Cluster Scheduler, and Platform LSF.
Notable features
- No single point of failure, backup daemons, fault-tolerant job options
- Highly scalable (schedules up to 100,000 independent jobs on the 100,000 sockets of IBM Sequoia)
- High performance (up to 1000 job submissions per second and 600 job executions per second)
- Free Software (GNU General Public License)
- Highly configurable with about 100 plugins
- Fair-share scheduling with hierarchical bank accounts
- Preemptive and gang scheduling (time-slicing of parallel jobs)
- Integrated with database for accounting and configuration
- Resource allocations optimized for network topology and on-node topology (sockets, cores and hyperthreads)
- Advanced reservation
- Idle nodes can be powered down
- Different operating systems can be booted for each job
- Scheduling for generic resources (e.g. Graphics processing unit)
- Real-time accounting down to the task level (identify specific tasks with high CPU or memory usage)
- Accounting for power usage by job
- Support of IBM Parallel Environment (PE/POE)
Coming in SLURM version 2.6 (2Q 2013)
- Support for job arrays
- Higher scalability
- Integration with Apache Hadoop + Open MPI based job launch
Coming in SLURM version 2.7 (4Q 2013)
- Hot spare nodes and other fault tolerance enhancements for long running jobs
- Energy efficient scheduling (including job specification of CPU frequency)
- Integration with FlexNet Publisher (FlexLM License Manager)
Supported platforms
While SLURM was originally written for Linux, the latest version supports many other operating systems:[2]
SLURM also supports several unique computer architectures including:
- IBM BlueGene L, P and Q models including the 20 petaflop IBM Sequoia
- Cray XT, XE and Cascade
- Tianhe-IA a 2.5 petaflop system with 14,336 Xeon processors and 7,168 NVIDIA GPUs
- Tera 100 a 1.25 petaflop Groupe Bull system with 140,000 Xeon cores
- Anton
Commercial support
In 2010, the developers of SLURM founded SchedMD, which maintains the canonical source, provides development, level 3 commercial support and training services. Commercial support is also available from Bright Computing, Bull. Cray, and Science + Computing
References
- ^ Pascual, Jose Antonio; Navaridas, Javier; Miguel-Alonso, Jose (2009). "Effects of Topology-Aware Allocation Policies on Scheduling Performance". Job Scheduling Strategies for Parallel Processing. Lecture Notes in Computer Science 5798. pp. 138–144. doi:10.1007/978-3-642-04633-9_8. ISBN 978-3-642-04632-2.
- ^ SLURM Platforms
- Balle, Susanne M.; Palermo, Daniel J. (2008). "Enhancing an Open Source Resource Manager with Multi-core/Multi-threaded Support". Job Scheduling Strategies for Parallel Processing. Lecture Notes in Computer Science 4942. p. 37. doi:10.1007/978-3-540-78699-3_3. ISBN 978-3-540-78698-6.
- Jette, M.; Grondona, M. (June 2003). "SLURM: Simple Linux Utility for Resource Management". Proceedings of ClusterWorld Conference and Expo (San Jose, California).
- Layton, Jeffrey B. (5 February 2009). "Caos NSA and Perceus: All-in-one Cluster Software Stack". Linux Magazine.
- Yoo, Andy B.; Jette, Morris A.; Grondona, Mark (2003). "SLURM: Simple Linux Utility for Resource Management". Job Scheduling Strategies for Parallel Processing. Lecture Notes in Computer Science 2862. p. 44. doi:10.1007/10968987_3. ISBN 978-3-540-20405-3.
