Originally referring to a specific computer built in 1994, Beowulf is a class of computer clusters similar to the original NASA system. Originally developed by Thomas Sterling and Donald Becker at NASA, Beowulf systems are now deployed worldwide, chiefly in support of scientific computing. They are high-performance parallel computing clusters of inexpensive personal computer hardware. The name comes from the main character in the Old English poem Beowulf, which was bestowed by Sterling because the epic poem describes the Beowulf as having "thirty men's heft of grasp in the gripe of his hand."
A Beowulf cluster is a group of what are normally identical, commercially available computers, which are running a Free and Open Source Software (FOSS), Unix-like operating system, such as BSD, GNU/Linux, or Solaris. They are networked into a small TCP/IP LAN, and have libraries and programs installed which allow processing to be shared among them.
There is no particular piece of software that defines a cluster as a Beowulf. Commonly used parallel processing libraries include Message Passing Interface (MPI) and Parallel Virtual Machine (PVM). Both of these permit the programmer to divide a task among a group of networked computers, and collect the results of processing. Examples of MPI software include OpenMPI (OpenMPI) or MPICH (MPICH). There are additional MPI implementations available.
Provisioning of Operating System and other software for a Beowulf Cluster can be automated using software, Open Source Cluster Application Resources(OSCAR) for example. OSCAR installs on top of a standard installation of a supported GNU/Linux distribution on a cluster's head node.
The following is the definition of a Beowulf cluster from the original how-to which was published by Jacek Radajewski and Douglas Eadline under the Linux Documentation Project in 1998.
Currently there are a number of GNU/Linux distributions, and one BSD, that are designed for building Beowulf clusters. These include:
- ABC Linux, based on Ubuntu - Allows the building of Beowulf clusters automatically, either live or by installing the software in the front end. All nodes run diskless.
- MOSIX is geared toward computationally-intensive, IO-low, applications.
- ClusterKnoppix, based on Knoppix - Last updated August 31, 2004
- ParallelKnoppix, Also based on Knoppix - Last updated May 29, 2008
- PelicanHPC (Successor to parallelknoppix based on Debian Live) - Last updated January 12, 2010
- Rocks Cluster Distribution
- DragonFly BSD
- Bootable Cluster CD
- Quantian, Live DVD with scientific applications, based on Knoppix and ClusterKnoppix. Last updated February 26, 2006.
Full article ▸