[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: ITS#3950



--On Saturday, January 14, 2006 11:59 PM -0800 Quanah Gibson-Mount <quanah@symas.com> wrote:

If there is an X in the B/E/N column, it is defined.


B N E R

X X X 5,726.250 searches/second
X X   5,626.316 searches/second
X   X 5,637.446 searches/second*
X     5,764.666 searches/second*
  X X 7,987.427 searches/second
  X   8,356.041 searches/second

* - These rates had a very spiky graph, i.e., the rate at any given time
slice was wildly varying.  The other search sets were relatively smooth.

Pulling in Howard's changes to back-bdb and tpool.c, in addition to the WAKE listener fix, I now have the following results on 2.6:

B N E R
X X X 7,384.195 searches/second
 X X 8,624.164 searches/second

So these changes significantly impacted when BROKEN is defined, and had a small (but positive) effect on when BROKEN isn't defined. BROKEN not being defined still outperforms BROKEN by a bit, however. And the BROKEN undef test is bordering max throughput on my 100MB ethernet (if only I had gigabit.. sigh). :P

--Quanah

--
Quanah Gibson-Mount
Principal Software Developer
ITSS/Shared Services
Stanford University
GnuPG Public Key: http://www.stanford.edu/~quanah/pgp.html