OpenLava
| Stable release | 3.3
|
|---|---|
| Written in | C |
| Operating system | Linux |
| Platform | Linux_x86_64 |
| Size | 1.53MB(.tar File) |
| Type | Job Scheduler for Compute Cluster |
| License | GNU General Public License |
| Website | www |
OpenLava[1] is an open source workload job scheduling software for a cluster of computers. OpenLava was derived from an early version of Platform LSF.[2] Its configuration file syntax, API, and CLI have been kept unchanged. Therefore, OpenLava is mostly compatible with Platform LSF.
OpenLava was based on the Utopia research project at the University of Toronto.[3]
OpenLava has gained significant development after forking out from Platform Lava, particularly in recent years.
Since fall 2015, the OpenLava development effort has been focused on improving scheduler performance and efficiency for high throughput workloads. In May 2016, OpenLava was benchmarked on AWS to schedule 1 million jobs with 100,000 concurrent job slots.[4]
OpenLava is licensed under GNU General Public License v2
History
In 2007, Platform Computing (now part of IBM) released Platform Lava 1.0, which is a simplified version of Platform LSF 4.2 code, licensed under GNU General Public License v2. Platform Lava had no additional releases after v1.0 and was discontinued in 2011.
In 2008, former Platform Computing employee David Bigagli created OpenLava 1.0 by forking code from Platform Lava.
In June 2011, OpenLava 1.0 code was committed to GitHub.[5]
In January 2012, OpenLava 2.0 was released with feature enhancements and bug fixes.
In November 2013, OpenLava 2.2 was released.
In April 2015, OpenLava 3.0 was released.
In November 2015, OpenLava 3.1 was released.
In February 2016, OpenLava 3.2 was released.
In May 2016, OpenLava 3.3 was released.
Notable features
Notable OpenLava features include the following:
- Automatic failover for the scheduler service
- Automatic failover for job failures
- Schedule jobs to thousands nodes
- Dynamic load based scheduling
- Round-robin
- Fair-share
- Preemption
- Priorities
- Load balancing
- Parallel job scheduling
- Multiple simultaneous scheduling policies
- Run and dispatch time windows
- Job limits
- Job dependencies
- Job migration
- Scheduling based on custom defined resources
- Job arrays
- Job suspension/resumption
- Complete job history
- Interactive job support
- Dynamic node membership
Commercial support
In 2014, a number of former Platform Computing employees founded Teraproc Inc.,[6] which contributes development and provides commercial support for OpenLava.[7] Commercially supported OpenLava contains add-on features than the community based OpenLava project.[8]
See also
- List of grid computing middleware distribution
- List of free and open-source software packages
- LSF
- GNU Queue
References
- ^ Jeff Laton. "openlava – Hot Resource Manager". Admin Magazine.
- ^ "IBM Platform LSF".
- ^ "Utopia: A Load Sharing Facility for Large, Heterogeneous Distributed Computer Systems". John Wiley & Sons. Retrieved 2007-12-29.
- ^ "OpenLava 3.3 – Benchmarking one million jobs on a 100,000 core cluster".
- ^ "openlava (openlava project)". GitHub. Retrieved 2016-03-21.
- ^ "An Old Platform Finds New Life Outside IBM Walls".
- ^ "Teraproc OpenLava Enterprise Edition".
- ^ "OpenLava Community Edition vs. OpenLava Enterprise Edition".