On Fri, 26 Mar 2010, Erik Trimble wrote:
It will attempt to balance the data across the two vdevs (the mirror and raidz) until it runs out of space on one (in your case, the mirror pair). ZFS does not currently understand differences in underlying hardware performance or vdev layout, so it can't "magically" decide to write data to one particular vdev over the other. In fact, I can't really come up with a
Zfs can see the backlog of writes to devices in a vdev (service time). It can also see how much free space there is in a vdev. Round-robin is also used to influence which vdev is used for the next write. Regardless, if data is written slowly it may be that the free space available in the vdevs will ultimately become the dominant factor.
Zfs may work just fine if the vdevs are of different base types, or use a different type of disk. It is just not assured to work as well as if the storage is perfectly balanced and symmetrical.
Maybe someone should experiment with this and report their findings. Bob -- Bob Friesenhahn bfrie...@simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/ GraphicsMagick Maintainer, http://www.GraphicsMagick.org/ _______________________________________________ zfs-discuss mailing list zfs-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
