On Sun, 11 Oct 1998, Andy Poling wrote:
> For RAID levels 0, 4 and 5 (the various "striping" RAID levels, it depends
> upon your goal.
>
> If you want to get maximum throughput for a single process by using all your
> spindles in parallel for each read or write operation, then you want a small
> chunk size.
If the algorithm of software RAID 0 is to distribute evenly (instead of
random) among the disks, then the ideal size of the chunk size is the
average size of data divided by the number of disks. It really depends on
the situation, if I have large files that are to be stored/accessed then
I would prefer larger chunks that are close to the ideal size of the
chunk.
>
> If, on the other hand, you want to randomize your spindle load (i.e. ideally
> have a different process reading or writing to/from each spindle), then you
> want a large chunk size. Some people also call this overlapped read and
> write.
You mean you're going to spawn several processes, just to read
simultaneously from the RAID device. I thought the best way is to use
RAID software to access the data from the devices concurrently and not
to worry about creating processes of your own to read on every disk.
>
> I don't think I understand how the concept of chunk size applies to RAID 1.
> Since it's supposed to fill one partition and then another, I don't see
> where chunk size comes into it... could just be me. :-)
Well, if you have larger chunks, then you can minimize disk access.
