Diskless shared-root cluster: Difference between revisions

Revision as of 13:38, 20 April 2008

Latest revision as of 15:51, 9 December 2021

(21 intermediate revisions by 15 users not shown)

Line 1:

{{no footnotes|date=August 2016}}

A '''Disklessdiskless Sharedshared-root Clustercluster''' is a way to easyly managedmanage several machines at the same time. Instead of each having its own [[operating system]] (OS) on its local disk, there is only one image of the OS available on a server, and all the nodes use the same image. (SSI cluster = Single[[single-system System Imageimage]])

The simplest way to achieve this is to use a NFS server, configured to host the generic boot image for the SSI cluster nodes. (dhcppxe + tftpdhcp + bootp or pxetftp + nfs)

To ensure that there is no [[Singlesingle point of failure]], the NFS export for the boot-image should be hostethosted on a two node cluster.

The architecture of a [[diskless]] [[computer cluster]] makes it possible to separate servers and storage array. The operating system as well as the actual reference data (userfiles, databases or websites) are stored competitively on the attached storage system in a centralized manner. Any server that acts as a cluster node can be easily exchanged by demand.

Line 9 ⟶ 10:

The additional abstraction layer between storage system and computing power eases the scale out of the infrastructure. Most notably the storage capacity, the computing power and the network bandwidth can be scaled independent from one another.

A similar technology can be found in the [[TruClusterVMScluster]] ([[Tru64OpenVMS]]-UNIX) in theand [[UnixTruCluster]] sector([[Tru64 UNIX]]).

A Linux cluster filesystem, e.g. [[Global File System|GFS]] or [[OCFS2]], is the basis to form a [[single-system image]] (SSI) on filesystem level with any attached [[Storage Area Network]] (SAN).

The [[open-source license|open-source]] implementation of a Disklessdiskless Sharedshared-root Clustercluster is known as [[Open-Sharedroot]].

Although such NFS based Diskless Shared-root Cluster are not recommended for productive environments, NFS still could be used to build cheap cluster infrastructures for less important tasks.

A similar technology can be found in the [[TruCluster]] ([[Tru64]]-UNIX) in the [[Unix]] sector.

The [[open-source]] implementation of a Diskless Shared-root Cluster is known as [[Open-Sharedroot]].

== Literature ==

* Marc Grimme, Mark Hlawatschek, Thomas Merz: [https://web.archive.org/web/20070126073113/http://www.redhat.com/magazine/021jul06/features/gfs_update/] ''Data sharing with a Red Hat GFS storage cluster''

* Marc Grimme, Mark Hlawatschek [https://web.archive.org/web/20071008115939/http://www.atix.de/downloads/shared-root-cluster.pdf] ''German Whitepaper: Der Diskless Shared-root Cluster'' (PDF-Datei; 1,1 MB)

* Kenneth W. Preslan: [http://www.redhat.com/docs/manuals/csgfs/admin-guide/] ''Red Hat GFS 6.1 – Administrator’s Guide''

==References==

== External links ==

<references/>

* http://www.open-sharedroot.org/ - The homepage of Open-Sharedroot

* http://sourceforge.net/projects/open-sharedroot/ - The project page of Open-Sharedroot

* http://w3.biff.ch/node/465/ - Tutorial how to implement Open-Sharedroot

* http://www.open-sharedroot.org/documentation/demo-cluster - Ready-to-run cluster within a VMware environment

{{DEFAULTSORT:Diskless Shared Root Cluster}}

[[Category:Cluster computing|Cluster, Diskless Shared Root]]

[[Category:Parallel computing|Cluster, Diskless Shared Root]]

[[Category:LocalComputer area networks|Cluster, Diskless Shared Rootnetworking]]