Diskless Shared Root Cluster

A diskless shared root cluster is a system platform for high-availability infrastructure.

A Linux cluster file system as GFS and OCFS2, is the foundation to connect a tethered Storage Area Network (SAN ) to a Single System Image ( SSI) at the file system level.

The architecture of a diskless computer cluster allows the separation of server system and storage array. Both the operating system, as well as the actual user data ( such as files, databases or websites ) are centrally maintained and competitively on the attached storage system. An acting as a cluster node server can be easily replaced if necessary.

Although there is a potential bottleneck, a cluster for simpler tasks can also be built without SAN by the data are stored on an NFS server. However, it is recommended that the NFS server itself is implemented as a 2- node cluster to avoid a single point of failure ( SPOF ).

Through the abstraction between the storage system and computing power of the server infrastructure scales well. Storage capacity, processing power and network capacity can be adjusted independently from one another to the current requirements.

A similar technology is in the UNIX area to find for example in TruCluster ( Tru64 UNIX).

An open- source implementation of a Diskless Shared Root Cluster is developed by the Open - Sharedroot project.