What is DFS, Distributed File System or System Distributed File?

Today, we are going to see what exactly DFS is. It is accessible in Microsoft environment under Windows Server. And the DFS acronym stands for Distributed File System or System Distributed File.

This system allows hierarchical file structure for shared files on different servers of logical network. It is used to reference a set of shares that will make available uniformly and to centralize all the space available on this set of shares.

With DFS, the end user does not display the name of the server on which it access read data, it is completely transparent. The advantage is that if the server has to change because of a failure or due to the developmental work, the path will remain the same.

Behind the same DFS path we can hide multiple servers with the same data with synchronization between these servers through DFSR (i.e. DFS Replication). This approach is very useful for high-availability of data and load balancing.

Of course, it is possible to use a path to a server (1 1), where each server will host different data.

Know that the DFS is increasingly present, now it is even used for SYSVOL replication on a domain between domain controllers.

Prerequisites:

Active directory and DNS are two important prerequisites for DFS; these two are core requirements for DFS with replication and network connectivity throughout DFS root.

Root folder and target

When speaking about DFS, three terms are important to remember and understand the workflow: DFS root, folder and target.

• DFS root: The main entrance level point of a DFS is it contains the path to the different DFS links connected with it. Additionally, there are two types of DFS roots.

• Folder: The folder will be the name displayed on the client side and the shared server configuration; a connection will be made between the DFS and DFS folder target to make connection between these two. Some files don’t use target which exclusively prioritizes DFS namespaces. The files are also called as “DFS Link”.

• Target: The server on which the data is stored. The target is the path to the shared folder on the server.

These three elements are essential to the establishment of a DFS infrastructure.

Architecture

To illustrate the point and for a better understanding for host DFS works, here is a brief example of DFS infrastructure:

On this infrastructure, our DFS server is “RV-AD01, which is also a domain controller in the domain “ESDS” and DNS. The DFS root is “ESDS”.

This DFS root contains two subfolders:

-Computer: which contains no target; there is only an organizational aspect of the namespace. This same contains two sub-folders “Data” and “Resources” that target “SRV01”; this server will be the one hosting the data for these two DFS folders.

-Work: contains two targets: SRV01 and SRV02, which will aim to host the data and ensuring high availability through replication via DFSR.

Benefits of DFS

The use of DFS enterprise has several important advantages.

• Simplify administration: if a target fails DFS, DFS link can be moved to another server containing a copy of the data by changing the target folder on the DFS server. The user side, it will be completely transparent because the name will not change.

• The DFS client is built into Windows which does not require additional installation on client computers.

• A unique name used to access all resources, it is not necessary to map a letter on each resource.

• Function caching to improve performance.

• The DFS considers the ACL located at the file system level.

• Replacement of a simplified server because the client side using namespace is not affected.

• Load Balancing (if several targets by DFS folder).

• Fault tolerance (if several targets by DFS folder).

• Evolution: Additional disk space can be added if the current disk space is not enough.