Re: Veritas's implementation/interpretation of RAID 0+1 and RAID 10

In comp.sys.sun.admin Darren Dunham <ddunham_at_redwood.taos.com> wrote:
> In comp.unix.solaris Chuck Swiger <chuck_at_codefab.com> wrote:
[ ... ]
> Regardless of the different failure performance of 0+1 and 1+0, the data
> remains in the same position on the platters. I've never noticed any
> performance differences between the two, so I was questioning why the
> different lists of applications.

I tend to agree with you, at least for the case where all of the disks are available. I suspect the situation would be different if you have a disk failure, or even multiple failures (given a large enough set of disks).

Also note that what you said here:

> Right. Although the setup makes it appear to be 0+1, it's really doing
> 1+0 under the covers.

...suggests that DiskSuite is smart enough to understand and perhaps even load-balance data available on multiple drives, which would explain why the same reads and writes happen to all of the disks, regardless of 0+1 vs. 1+0.

-Chuck

       Chuck Swiger | chuck_at_codefab.com | All your packets are belong to us.
       -------------+-------------------+-----------------------------------
       "The human race's favorite method for being in control of the facts
        is to ignore them."  -Celia Green