On Wed, Jul 23, 2014 at 10:47 PM, Prakash Surya <[email protected]> wrote:

> Out of curiosity, is the workload used to measure the 18% performance
> benefit as simple as it seems? i.e. fill the arc, then do reads which
> are guaranteed to be arc hits?
>

That's right.

--matt


>
> --
> Cheers, Prakash
>
> On Wed, Jul 23, 2014 at 10:06:09PM -0700, Matthew Ahrens wrote:
> >    Good point, it should be consistent with the others.  Review updated.
> >    --matt
> >
> >    On Wed, Jul 23, 2014 at 9:50 PM, Prakash Surya <[1]
> [email protected]>
> >    wrote:
> >
> >      Why not prefix the variable with "zfs_" like the rest of the
> >      parameters?
> >      Seems reasonable otherwise, I've seen large hash lengths on the
> >      Linux
> >      port running Lustre-ZFS MDS servers which makes use of many 512B
> >      buffers.
> >      --
> >      Cheers, Prakash
> >      On Wed, Jul 23, 2014 at 08:36:29PM -0700, Matthew Ahrens wrote:
> >      >    [1][2]https://reviews.csiden.org/r/61/
> >
> >    >    5034 ARC's buf\_hash\_table is too small
> >
> >      >    Reviewed by: George Wilson <[2][3][email protected]\>
> >      >    Reviewed by: Christopher Siden
> >      <[3][4][email protected]\>
> >
> >    >    Original author: Matthew Ahrens
> >    >    The ARC puts all (non-anonymous) arc_buf_hdr_t's in a hash table,
> >    which
> >    >    is
> >    >    created at system boot time.  The hash table is sized such that
> if
> >    all
> >    >    of
> >    >    physical memory was filled with 64K blocks, the hash chain length
> >    would
> >    >    average
> >    >    less than 1.0.  However, on a system with typical block size of
> >    8k,
> >    >    this can lead to
> >    >    long hash chain lengths.  I've observed average length ~6.5;
> >    >    theoretically it
> >    >    could be up to 16 (because evicted "ghost" entries are also in
> the
> >    >    hashtable).
> >    >    By increasing the hash table size to have enough entries for
> >    average
> >    >    length 1.0
> >    >    when memory is filled with 8k blocks, we can obtain a 18%
> >    performance
> >    >    improvement on cached reads.  (680MB/s -> 805MB/s)
> >    >    The hash table size should also be tunable, rather than hard
> >    coded.
> >    >    --matt
> >    >
> >
> >      > References
> >      >
> >      >    1. [5]https://reviews.csiden.org/r/61/
> >      >    2. mailto:[6][email protected]
> >      >    3. mailto:[7][email protected]
> >      > _______________________________________________
> >      > developer mailing list
> >      > [8][email protected]
> >      > [9]http://lists.open-zfs.org/mailman/listinfo/developer
> >
> > References
> >
> >    1. mailto:[email protected]
> >    2. https://reviews.csiden.org/r/61/
> >    3. mailto:[email protected]
> >    4. mailto:[email protected]
> >    5. https://reviews.csiden.org/r/61/
> >    6. mailto:[email protected]
> >    7. mailto:[email protected]
> >    8. mailto:[email protected]
> >    9. http://lists.open-zfs.org/mailman/listinfo/developer
>
> > _______________________________________________
> > developer mailing list
> > [email protected]
> > http://lists.open-zfs.org/mailman/listinfo/developer
>
>
_______________________________________________
developer mailing list
[email protected]
http://lists.open-zfs.org/mailman/listinfo/developer

Reply via email to