FAQ's, How-to's, Guides
Tutorial: Building a Beowulf System
by: Jan Lindheim, Caltech, (May 2005)
With the power and low prices of today's off-the-shelf PCs and the availability of 100 Mb/s ethernet interconnect, it makes sense to combine them to build High-Performance-Computing and Parallel Computing environment. This is the concept behind the Beowulf parallel computing system we will describe. With free versions of Unix and public domain software packages, no commercially available parallel computing system can compete with the price of the Beowulf system. The drawback to this system is, of course, there will exist no support center to call when a problem arises. But there does exist a wealth of good information available through FTP sites, web sites and newsgroups.
Engineering a Beowulf-style Compute Cluster
by: Robert G. Brown, Duke University
This is a local copy of my online book on beowulf style cluster engineering. It is (and will likely always be) a work in progress, so check the revision number and dates from time to time to see if new material has been added.
Redhat's Linux Cluster Project Pages
GFS | CLVM | CCS | CMAN | DLM | GULM | Fence | GNBD | CSNAP | LVM2 | Device-Mapper
Diskless, NFS, OpenMosix HOWTO
A simple step by step non distro specific HOWTO to
setup a OpenMosix cluster. Using Wake on LAN, NFS,
TFTP, and diskless clients. A couple of unique parts
include a reboot script, to reboot a node by pressing
the space bar, so you can use a PC as a regular
workstation in the day and as a cluster node at night,
if the BIOS supports network boot on WOL and hard
drive boot otherwise (newer Dells do). Also the MP3
ripping script will count the number of nodes
currently up and start the same number of rips.
Cluster Quick Start (DRAFT)
There are many ways to configure a cluster. This document describes one way. This guide not complete, it is really just a series of steps or issues that need to be addressed before a cluster can be usable. Indeed, it is very difficult to create a "step by step" set of instructions because every cluster seems to
be different.
Beowulf Installation and Administration HOWTO (0.0.6)
Jacek Radajewski started writing the Beowulf HOWTO in November 1997 and was soon joined by Douglas Eadline. Over a few months the Beowulf HOWTO grew into a large document, and in August 1998 it was split into three documents: Beowulf HOWTO, Beowulf Architecture Design HOWTO, and the Beowulf Installation and Administration HOWTO. Version 1.0.0 of the Beowulf Installation and Administration HOWTO will be released to the Linux Documentation Project soon.
Beowulf HOWTO (1.1.1)
Jacek Radajewski started work on this document in November 1997 and was soon joined by Douglas Eadline. Over a few months the Beowulf HOWTO grew into a large document, and in August 1998 it was split into three documents: Beowulf HOWTO, Beowulf Architecture Design HOWTO, and the Beowulf Installation and Administration HOWTO. Version 1.0.0 of the Beowulf HOWTO was released to the Linux Documentation Project on 11 November 1998. We hope that this is only the beginning of what will become a complete Beowulf Documentation Project.
COCOA Beowulf Cluster FAQ (1.1)
COCOA stands for COst effective COmputing Array. It is a Beowulf class supercomputer. Beowulf is a multi computer architecture which can be used for parallel computations. It is a system which usually consists of one server node, and one or more client nodes connected together via Ethernet or some other fast network. It is a system built using commodity hardware components, like any office desktop PC with standard Ethernet adapters, and switches. It does not contain any custom hardware components and is trivially reproducible.
FNN: Flat Neighborhood Networks
Welcome to the home of FNN documents and software! Since the first press release on our cluster KLAT2, which has a Flat Neighborhood Network, many of you have been asking us for more information, access to the GA (genetic search algorithm) we developed to design FNNs, etc. This is where everything will be posted....
Linux Cluster HOWTO - How to set up high-performance Linux computing clusters.
YABIH - Yet Another Beowulf Installation Howto
This is a very short and somewhat old Howto. It describes a few steps in setting up a cluster.
Clusters Docs
The aim of this page is to collect documentation about how to set up, maintain, program and use a Linux cluster for high-performance computing. Actually, there are many web pages with documentation about this topic: tutorials, HOWTOs, forums, and so on. These pages can be helpful to answer frequently asked questions and to solve some common problems when working with Linux clusters; however, there isn't any attempt to collect all this documentation into a single source.
A Survey of Cluster Technologies (High Availability)
This paper surveys the cluster technologies for the operating systems available from many vendors, including IBM AIX, HP HP-UX, Linux, HP NonStop Kernel, HP OpenVMS, PolyServe Matrix Server, Sun Microsystems Solaris, HP Tru64 UNIX and Microsoft Windows 2000/2003.
IBM: Linux HPC Cluster Installation
This redbook will guide system architects and systems engineers toward a basic understanding of cluster technology, terminology, and the installation of a Linux High-Performance Computing (HPC) cluster (a Beowulf type of cluster) into an IBM eServer xSeries cluster.
