Re: Are large RAID stripe sizes useful with FreeBSD?

From: Ivan Voras <ivoras_at_freebsd.org>
Date: Mon, 31 Mar 2008 21:51:20 +0200
On 31/03/2008, Xin LI <delphij_at_delphij.net> wrote:
> Ivan Voras wrote:
>  > Most of new hardware RAID controllers offer stripe sizes of 128K, 256K
>  > and some also have 512K and 1M stripes. In the simplest case of RAID0 of
>  > two drives, knowing that the data is striped across the drives and that
>  > FreeBSD issues IO request of at most 64K, is it useful to set stripe
>  > sizes to anything larger than 32K? I suppose something like TCQ would
>  > help the situation but does anyone know how is this situation usually
>  > handled on the RAID controllers?
>
>  Short answer: yes.  Larger stripe sizes would allow better performance
>  for continuous read/write.  In theory they will negatively affect small
>  size read/write but in practice you will notice that larger stripe size
>  would give better performance in almost all cases.

Ok, I'm interested in how it works. Are the IO requests queued at the
controller until there's enough of them to write a stripe block?
Received on Mon Mar 31 2008 - 18:19:06 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:29 UTC