RE: raid0 vs. raid5 read performance

[Tom Livingston]

Ingo Molnar wrote:
> My experience is that RAID5
> read-performance is almost as high as RAID0 performance when using 4k
> stripe size. In this case both Linux and the disk itself has more chances
> to optimize. (disks will most likely read sequentially due to readahead
> caching, and they will skip over parity blocks without skipping
> physically). Nevertheless this means RAID5 performance will never be
> better than (N-1)/N*RAID0_bandwith.

Are there drawbacks to setting your stripe size to be 4k ? The FAQ I used
when setting up my 10 drive raid5 array suggested 128k as a starting point,
which is what I used.

Others on this list seem to be confused as to how to calculate an optimum
stripe size.  Would this be a good thing to "get the word out" on, or is
there a more appropriate way to determine it?

Tom