Re: stochastic fair queueing in the elevator [Re: [BENCHMARK] 2.4.20-ck3 / aa / rmap with contest]

[Andrea Arcangeli]

On Mon, Feb 10, 2003 at 11:17:30PM +1100, Nick Piggin wrote:
> Andrea Arcangeli wrote:
> 
> >On Mon, Feb 10, 2003 at 03:48:08AM -0800, Andrew Morton wrote:
> >
> >>Andrea Arcangeli <[e-mail]> wrote:
> >>
> >>>It's the readahead in my tree that allows the reads to use the max scsi
> >>>command size. It has nothing to do with the max scsi command size
> >>>itself.
> >>>
> >>Oh bah.
> >>
> >>-               *max_ra++ = vm_max_readahead;
> >>+               *max_ra = ((128*4) >> (PAGE_SHIFT - 10)) - 1;
> >>
> >>
> >>Well of course that will get bigger bonnie numbers, for exactly the 
> >>reasons
> >>I've explained.  It will seek between files after every 512k rather than
> >>after every 128k.
> >>
> >
> >NOTE: first there is no seek at all in the benchmark we're talking
> >about, no idea why you think there are seeks. This is not tiobench, this
> >is bonnie sequential read.
> >
> Yes, Andrew obviously missed this... Anyway, could it be due to
> a big stripe size and hitting more disks in the RAID? How does
> a single SCSI disk perform here, Andrea?

Actually I increased readahead to more than just 512k in my last tree,
especially to take care of RAID :) so that both lowlevel will get the
512k, instead of being limited to 256k command each 8) This might apply
to hardware raid too.

Andrea