I'm one of those people who use the RAID system for replicating data for offsite backup. It works, but is a kludge.
I've just discovered a newly released project that has piqued my interest as a possible replacement. It is called Zumastor and looks like it might be an efficient system for block replication on Linux. http://zumastor.org/ Still only version 0.6, so maybe only something to watch. Assuming write performance is not badly hit when snapshotting (otherwise normal BackupPC operation will be slowed down - see last point in this email), I think the mode of operation for server backup would be. Stop BackupPC Take new snapshot Start BackupPC Replicate changes between new snapshot and previous snapshot to other system (I wonder if it can replicate to another local disk?) Remove old snapshot (but leave new snapshot around for next replicate operation) The SCALE talk mentions the possibility for multi-master which might be interesting for those running multiple servers backing up the same clients - a kind of backup server cluster might be possible. The feature that makes the difference, and saves the "seek all around the disk to find new/updated files since last backup" problem is the fact that changed blocks are tracked as they are modified. This problem of course still exists between the BackupPC server and the clients though. SiMoN ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ BackupPC-users mailing list BackupPC-users@lists.sourceforge.net List: https://lists.sourceforge.net/lists/listinfo/backuppc-users Wiki: http://backuppc.wiki.sourceforge.net Project: http://backuppc.sourceforge.net/
