Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

First, I am open to suggestion on the title of this.

Here is what I am looking for: I am rebuilding our file server because it’s dying, as well as space being poorly allocated. What I would like to be able to do, is store data accessed in the last year near line, and to replicate that data, using Windows DFS, while storing any unused data on a cheaper and slower storage medium, and the only redundancy would be tape backup.

So, for instance, I have a total of about 5TB of storage capacity, but only the most active 1TB I would keep near line and would replicate using Windows DFS. Leaving the rest available, but in a disaster, needing to be recovered from tape. The older data would be what would be considered important to keep, but not mission critical for day to day operations.

What I would also like to do would be to seamlessly "stitch" this data together so it continues to look like it’s still in the same folder, and let windows figure out where the data is actually located.

SQL server 2008 has a good example of this with filegroups, where I can store last year’s data on slower cheaper SATA drives, and SQL will seamlessly stitch the entire database together for me, so as a user I never know that the data I am looking at infrequently is stored somewhere else.

I am using Windows 2008 R2 standard with the file services role installed. I would prefer to keep this as a native solution without getting into 3rd party software, but I am not shutting the door completely on it.

share|improve this question
    
have not fully investigate MS's FCI process but it may do what your looking for. –  tony roth Jul 15 '10 at 18:05
add comment

2 Answers

up vote 1 down vote accepted

Actually now that I've read up on MS's FCI, I'm sure that task is doable!

http://blogs.technet.com/b/filecab/archive/2009/05/11/windows-server-2008-r2-file-classification-infrastructure-managing-data-based-on-business-value.aspx

share|improve this answer
    
This looks to be pretty close to perfect for what I need. –  Russ Jul 15 '10 at 18:37
    
I may have found a kink in the works, I don't think it will do the "stich" component. Still looking into this. –  tony roth Jul 15 '10 at 18:53
    
via a script with mklink this can be accomplished! –  tony roth Jul 21 '10 at 16:38
add comment

What you're taking about is Hierachical Strorage Management with a Unified Name Space - it's what people like EMC, NetApp and Hp are all very good at. Basically you have the ability to 'see' all of your data ever created but only the most used data is very fast and resilient with lesser data slower and less resilient. So solutions for this exist, I know - I have a lot of these systems...but you want this 'in OS', which will require other SF users far greater Windows skills than mine. But keep ib mind that you can get these systems to work if the OS-driven route dries up, just not without a price tag ok.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.