Clustered File System

Clustered file system, as the name suggest, is a file system wherein the entire system is shared by mounting the same over multiple servers. Means, several servers use a single file system, thus the name “clustered”.

There are several approaches for clustering file system. It is interesting to note, however, that not all clustering methods may make use of clustered file system.

The clustered file system offers multiple objectives at once. First, it makes location-independent addressing. Second, it allows for redundancy in the system. By making the system redundant, you can reduce the overall complexity of the system and make it more efficient.

Parallel file system is a form of clustered file system which deploys several nodes and distributes data across them at once. This not only involves load sharing but also the encryption of data in the system. A shared-disk file system is the best example of clustered file system that allows several computers at once to gain direct access to the pieces at the block level.

Access control and translation always takes place at the client node and never at the system end. Thus us because any upcoming requests must be checked right at the network entry. By implementing system check at the server node you are only making the system more vulnerable to attack. Thus, it is the best practice to implement access control in the user node in clustered file system.