Sun Grid Engine
From Wikipedia, the free encyclopedia
Sun Grid Engine |
|
---|---|
Developed by | Sun Microsystems in association with the community |
Latest release | 6.1 update 4 / April 4, 2008 |
Preview release | 6.2 Beta / May 13, 2008 |
OS | Cross-platform |
Genre | Grid computing |
License | SISSL |
Website | gridengine.sunsource.net |
Sun Grid Engine (SGE), previously known as CODINE (COmputing in DIstributed Networked Environments) or GRD (Global Resource Director),[1] is an open source batch-queuing system, supported by Sun Microsystems. Sun also sells a commercial product based on SGE, also known as N1 Grid Engine (N1GE).
SGE is typically used on a computer farm or computer cluster and is responsible for accepting, scheduling, dispatching, and managing the remote execution of large numbers of standalone, parallel or interactive user jobs. It also manages and schedules the allocation of distributed resources such as processors, memory, disk space, and software licenses.
SGE is the foundation of the Sun Grid utility computing system, made available over the Internet in the United States in 2006,[2] later becoming available in many other countries.
Contents |
[edit] Features
Features of SGE 6.1 include:
- Multiple advanced scheduling algorithms allow powerful policy-based resource allocation
- Cluster queues
- Job and scheduler fault tolerance
- Job checkpointing
- Job arrays and job tasks
- DRMAA (Job API)
- Resource reservation
- XML status reporting (qstat and qhost), and the xml-qstat web interface
- Parallel jobs (MPI, PVM, OpenMP), and scalable parallel job startup with qrsh [3]
- Usage accounting
- accounting and reporting (ARCO)
- parallel make: distmake, dmake (Sun Studio), and SGE's own qmake
- FLEXlm integration [4] and multi-cluster software license management with LicenseJuggler [5]
[edit] Upcoming features in version 6.2
- Advance reservation
- Array job interdependencies
- Enhanced remote execution (without using external rshd/rlogind/sshd processes)
- Multi-clustering [6]
- Daemons managed by the Service Management Facility on Solaris
[edit] Platforms
SGE runs on multiple platforms, including:
- AIX
- BSD - FreeBSD, NetBSD, OpenBSD
- HP-UX
- IRIX
- Linux
- Mac OS X
- Solaris
- SUPER-UX
- Tru64
- Windows (as execution hosts only)
- Z/OS (in progress)
[edit] Support and training
Sun provides support contracts [7] for the commercial version of Grid Engine on most UNIX platforms and Windows. Professional services, consulting, training, and support are also provided by Sun Partners. [8] Sun partners with Georgetown University to deliver Grid Engine administration classes.[9] The Bioteam runs short SGE training workshops that are 1 or 2 days long.[10]
Users can get community support on the Grid Engine mailing lists.[11]
Grid Engine Workshops were held in 2002, 2003, and 2007 in Regensburg, Germany.[12]
[edit] Prominent users
Notable deployments of SGE include:
- Sun Grid
- the TSUBAME supercomputer at the Tokyo Institute of Technology,[13] which was number 7 on June 2006 TOP500 list.
- Ranger at the Texas Advanced Computing Center (TACC). Ranger has 62,976 processor cores in 3,936 nodes and a peak performance of 504TFlops. [14] [15]
[edit] History
In 2000, Sun acquired Gridware, Inc. a privately owned commercial vendor of advanced computing resource management software with offices in San Jose, Calif., and Regensburg, Germany.[16] Later that year, Sun offered a free version of Gridware for Solaris and Linux, and renamed the product Sun Grid Engine.
In 2001, Sun made the source code available,[17] and adopted the open source development model. Ports for Mac OS X and *BSD were contributed by the non-Sun open source developers.
[edit] Other Grid Engine based products
- Sun Constellation System
- Sun Visualization System
- Rocks Cluster Distribution
- EGEE [18]
- Univa UD's Cluster Express
- BioTeam's iNquiry
[edit] Add-on software
A number of SGE add-ons are available:
- Solaris Cluster integration [19]
- Service Domain Management module in order to meet service level objectives
- Transfer-queue Over Globus (TOG). Globus does not officially support SGE, but the London e-Science Center has created a package and provides instructions on how to configure a Globus Toolkit server so that it can submit jobs for execution on a local Sun Grid Engine installation.
- JOb Scheduling Hierarchically (JOSH)
[edit] References
- ^ A Little History Lesson. Sun Microsystems (2006-06-23).
- ^ World's First Utility Grid Comes Alive on the Internet. Sun Microsystems (2006-03-22).
- ^ Long delay when submitting large jobs (mailing list message). Sun Microsystems. Retrieved on 2007-12-25.
- ^ Olesen-FLEXlm-Integration. wiki.gridengine.info. Retrieved on 2007-12-25.
- ^ LicenseJuggler. wiki.gridengine.info. Retrieved on 2007-12-26.
- ^ Hedeby Project home. Sun Microsystems. Retrieved on 2008-01-25.
- ^ Sun Store Grid Engine Entitlement Purchase. Sun Microsystems. Retrieved on 2008-03-03.
- ^ Sun Grid Engine 6 Partners. Sun Microsystems. Retrieved on 2007-12-14.
- ^ Advanced Sun Grid Engine Configuration and Administration Class. Sun Microsystems. Retrieved on 2007-12-14.
- ^ Training. The Bioteam Inc.. Retrieved on 2008-03-24.
- ^ Grid Engine Mail Lists. Sun Microsystems. Retrieved on 2008-01-23.
- ^ Grid Engine Workshops. Sun Microsystems. Retrieved on 2007-12-14.
- ^ Sun N1 Grid Engine Software and the Tokyo Institute of Technology Super Computer Grid. Sun Microsystems. Retrieved on 2007-11-16.
- ^ TACC > HPC Systems. The University of Texas at Austin. Retrieved on 2007-12-13.
- ^ More Ranger Facts and Figures. Sun Microsystems. Retrieved on 2008-02-12.
- ^ Gridware's resource management software increases efficiency and productivity in compute-intensive technical computing environments. Sun Microsystems (2000-07-24).
- ^ Sun Microsystems makes SUN GRID ENGINE software available to open source community. Sun Microsystems (2001-07-23).
- ^ Sun Grid Engine, a new scheduler for EGEE middleware. Imperial College (2000-12-29).
- ^ Installing and Configuring Sun Cluster HA for Sun Grid Engine. Sun Microsystems (2008-02-15).
[edit] See also
- Sun xVM - Sun's datacenter automation tool
- Open Grid Forum
[edit] External links
- gridengine: Home - Official project page
- Sun N1 Grid Engine - Official page for the commercial version
- Sun Grid Engine 6.0 Collection - Documentation Collection for Grid Engine 6.0
- Sun Grid Engine 6.1 Collection - Documentation Collection for Grid Engine 6.1
- Sun Grid Engine Information Center - Documentation Collection for Grid Engine 6.2
- gridengine.info - Unofficial Grid Engine resources blog
- wiki.gridengine.info - Unofficial Grid Engine resources wiki - including information about integration of applications
- "Understanding the differences between Grid Engine 5.3, 6.0 and Sun N1 Grid Engine 6 (N1GE 6)"
- xml-qstat project page - Web based status monitoring of Grid Engine 6.x systems
- Good Tips - several blog entries on getting started with Grid Engine
- Grid Engine-Globus Toolkit adapter
- BinBase Cluster System - an API to access the SGE from java and run java based jobs on the cluster.
- Schedule::SGE - An API to access the SGE using Perl
|