You might check with OSC on backing this stuff up, for which Doug Johnson 
would be a good first contact.  They have several hundred TB of mass 
storage for statewide research and archival use.

The *.mat format already involves some data compression, but it may be 
worth looking into compressing them further.

Steve Wright

At 11:09 AM 5/9/2007, you wrote:
>OK, here is what I learned from Amit.
>
>He is still generating these files, but hopes to start analyzing them on
>Friday.  Once he has finished with the analysis he can delete most of them,
>but would like to write them to tape, or some other permanent backup in case
>he needs to do more work with them.  He said that if needed he can
>re-generate the files, but it has taken almost three weeks to generate them,
>so he would rather not have to do that.
>
>Does this help with management of the backups?  Any thoughts on how to
>provide a long-term/permanent backup of the files (Robin has already
>suggested an external hard drive)?
>
>If you need any more info or want to make changes to how these files are
>backed up and want to make sure Amit is OK with them, let me know or just
>e-mail him directly.
>
>Dave
>
>-----Original Message-----
>From: Research Computing Support [mailto:[log in to unmask]] On
>Behalf Of Steve Moore
>Sent: Wednesday, May 09, 2007 9:55 AM
>To: [log in to unmask]
>Subject: Backing up Matlab files
>
>My short question is: do we need to backup matlab files?  Dr.
>Shukla's job (which I presume is matlab) has generated about 2
>terabytes of .mat files recently.  This is causing space and
>performance problems on our primary backup server.  If no one knows
>the answer I will contact Matlab.
>
>Steve Moore
>Senior Software Coordinator
>Miami University
>113 B Hoyt Hall
>Oxford, OH 45056
>email: [log in to unmask]
>voice: 513-529-1452
>fax: 513-529-9665