Post reply

Use of FileStream considering SQL Server backups

  • September 27, 2011 at 9:50 am

    #390486

    Hi,

    We have configured FileStream RBS for Share Point 2010 databases in SQL Server 2008 R2.

    We are using Native SQL Server backup to backup the database and this backup also taking the backup up blob files.

    So ultimately it's taking the same time for backups with or without using filestream RBS.

    I can see the database size is small but the backup files very large and time consuming.

    File Stream is basically for easy maintenance of databases but I'm not achieving that considering the backups. Could you please advice are there any better backup methods/tools?

    Thanks

  • October 5, 2011 at 9:56 am

    #1390789

    Could you please shed some light on this post?

    Thanks

  • October 5, 2011 at 10:06 am

    #1390794

    I'm not sure what the problem is here. Filestream data is backed up with the database, that's one of the key advantages of it over regular files in the filesystem.

    Gail Shaw
    Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
    SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability

    We walk in the dark places no others will enter
    We stand on the bridge and no one may pass

  • February 8, 2012 at 10:37 am

    #1444659

    On other systems, I have found that turning "Database Backup compression" helps a great deal...often up to 80% compression. Unfortunately, with these sharepoint collections, it only compresses maybe 5-10% percent, maybe not even enough to turn it on.

    RBS doesn't really help with a backup strategy. It is possible to do "partical backups" by filegroup, but that is not recommended. A normal Full native SQL backup is just merging the data and files back into one .bak so nothing is gained in terms of backups. If you did partial backups, separating your db backup and trying to use a more efficient file system backup on the RBS store(since the individual files are mostly static)...you would have so many consistency issues later if you restored from two separate backup systems. That seems to be the consensus from what I've found.

    Our best approach is just to always name our native sql backup files with a timestamp...so we don't reuse the name. We do a full once a week on sharepoint, which the backup system writes to tape once and then ignores etc. With most backup systems, like Legato/Tivoli, if you are doing filesystem backups of your native sql .bak and .trn backups then, you just want uniqueness in your naming convention so it picks up once. You don't want to use a backup device, appending etc...or you are backing up too much...since that file will be touched and picked up by your tape backup system each night etc.

Viewing 4 posts - 1 through 3 (of 3 total)

You must be logged in to reply to this topic. Login to reply