DFSR vs. a Global File System

Today we will be talking about the global file system versus DFSR.

In considering DFSR, it’s important to consider why we want to replicate files across multiple sites. If you have multiple file servers and you want to be able to produce a copy of the files from a source to a target, you will use a product like DFSR to do that. Now, the objective here is typically to separate geographically those two sites so that in the event that one of the sites goes down you still have a secondary site, a failover site, that you can restore to.

Now that means that this is providing some backup. It’s not a perfect solution for backup because it’s missing real clear versioning. It is providing some ability to DR because you can always fail over to the secondary site and it is providing, to some degree, business continuity because you can always fail the users that were in this site to the secondary site. 

Some of you may want to use the same technology as a way to collaborate across sites. This is where a lot of people get in trouble. Where they want to have more than one active file server at the same time. You get file collisions and the fact that you have no versioning, no protection, means that the users are very unhappy about loosing work when there is any kind of conflicts on the files.

By comparison a global file system allows you to set up a hub and spokes architecture where you can connect as many file servers as you want, and all those file servers are replicating against a central hub for the files. What that allows you to do is the global file system keeps a stream of all the versions of the files here centralized and there is constantly reconciling of the versions of the files against each of these sites, and the changes across all the sites.

This is very helpful when you want to collaborate, especially when coupled with a locking mechanism that guarantees that if a user is trying to change a file in this location, those changes cannot be made at the same time in this other location, or this other location, or any of the other locations.

Furthermore, if you couple this approach with something like DFSN across all of the sites, then what you get is instant failover, and business continuity across the sites, because any one of these sites could fail. And because the other sites are synchronizing continuously against this site the end user could just move smoothly and without any conflicts to a secondary site until you’ve had a chance to restore the original site. 

This is powerful technology. A global file system allows you to both collaborate and, in the simple case, to eliminate any need for backup or for DR and business continuity as a separate solution.