It is about backup windows, also. :) NDMP to TSM at $work is serially streamed for reasons I hear are TSM limitations. To reduce the backup window, we split data (VMware datastores) across several smaller flexvols as opposed to one flexvol to rule them all. IANABA, but it has to do with being able to concurrently go after drives in the library fo writes or something like that.
But RTO is also worth evaluating (with more weight too :). On Fri, Apr 30, 2010 at 11:30 AM, Travis <[email protected]> wrote: > On Fri, Apr 30, 2010 at 7:10 AM, Andrew Hume <[email protected]> wrote: > >> 6) i hadn't worried about backup time before, and still don't. backup time >> is about total amount to be backed up and in my experience, not too affected >> by FS >> size. > > I'd worry about restore time. In my experience, it was always 2-3 > times longer to > do restores than it was to do the backups on large filesystems. Some > of this may have > been driven by the speed of our restore media; we never really did > disk-to-disk. > > Invariably, it's during a DR event that you need to shave time, not during a > backup window. > > Travis > > -- > Travis Campbell > [email protected] > _______________________________________________ > Tech mailing list > [email protected] > http://lopsa.org/cgi-bin/mailman/listinfo/tech > This list provided by the League of Professional System Administrators > http://lopsa.org/ > _______________________________________________ Tech mailing list [email protected] http://lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
