> From: zfs-discuss-boun...@opensolaris.org [mailto:zfs-discuss- > boun...@opensolaris.org] On Behalf Of Trond Michelsen > > Hi.
I think everything you said sounds perfectly right. As for estimating the time required to "zfs send" ... I don't know how badly "zfs send" gets hurt by the on-disk order or fragmentation of data due to normal pool wear and tear. But there's a really easy way to find out. "time zfs send somepool | pv > /dev/null" The in-use pool might take a long time to send, depending on how it's physically laid out on disk. But when it's received once ... and subsequently sent & received again ... These operations should all be extremely efficient, and almost entirely limited by sustainable drive throughput instead of random seek time. The zfs send | receive operations are essentially defragging for you as you go along. _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
