netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephen Hemminger <shemminger@vyatta.com>
To: Octavian Purdila <opurdila@ixiacom.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>,
	Krishna Kumar2 <krkumar2@in.ibm.com>,
	Hagen Paul Pfeifer <hagen@jauu.net>,
	netdev@vger.kernel.org
Subject: Re: [PATCH next-next-2.6] netdev: better dev_name_hash
Date: Mon, 26 Oct 2009 09:55:16 -0700	[thread overview]
Message-ID: <20091026095516.02f1cb49@nehalam> (raw)
In-Reply-To: <200910261752.51784.opurdila@ixiacom.com>

Added more algorithms to test...

Time is in seconds for 10000000 entries with hashbits = 8
Ratio is number of probes / ideal hash probes

Result sorted by distribution:

Algorithm             Time       Ratio       Max   StdDev
string10             1.434087       1.00     39064   0.01
SuperFastHash        1.469511       1.00     40497   2.17
string_hash17        1.472544       1.00     39497   1.50
jhash_string         1.501508       1.00     39669   1.04
crc                  2.826795       1.00     39088   0.07
md5_string           3.608253       1.00     39605   0.98
djb2                 1.462722       1.15     60681  76.16
string_hash31        1.457253       1.21     64950  91.12
sdbm                 1.566174       2.38    129900 232.22
pjw                  1.527306       2.45     99990 237.86
elf                  1.576096       2.45     99990 237.86
kr_hash              1.400072       7.80    468451 515.52
fletcher             1.449671       7.80    468451 515.52
full_name_hash       1.487707      13.09    562501 687.24
xor                  1.400403      13.36    583189 694.98
lastchar             1.348798      25.60   1000000 980.27

Another run sorted by speed:
Algorithm             Time       Ratio       Max   StdDev
lastchar             1.338545      25.60   1000000 980.27
kr_hash              1.398453       7.80    468451 515.52
xor                  1.398843      13.36    583189 694.98
string10             1.432756       1.00     39064   0.01
fletcher             1.448499       7.80    468451 515.52
string_hash31        1.457524       1.21     64950  91.12
string_hash17        1.462548       1.00     39497   1.50
djb2                 1.462956       1.15     60681  76.16
SuperFastHash        1.469907       1.00     40497   2.17
full_name_hash       1.486465      13.09    562501 687.24
jhash_string         1.500959       1.00     39669   1.04
pjw                  1.526097       2.45     99990 237.86
sdbm                 1.566533       2.38    129900 232.22
elf                  1.576470       2.45     99990 237.86
crc                  2.811210       1.00     39088   0.07
md5_string           3.604675       1.00     39605   0.98


  reply	other threads:[~2009-10-26 16:55 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-10-25 19:58 [PATCH next-next-2.6] netdev: better dev_name_hash Octavian Purdila
2009-10-25 20:17 ` Hagen Paul Pfeifer
2009-10-25 21:24 ` Eric Dumazet
2009-10-25 21:55   ` Octavian Purdila
2009-10-25 22:41     ` Hagen Paul Pfeifer
2009-10-25 22:45       ` Octavian Purdila
2009-10-26  5:28       ` Eric Dumazet
2009-10-26 13:07         ` Krishna Kumar2
2009-10-26 14:31           ` Octavian Purdila
2009-10-26 14:55             ` Eric Dumazet
2009-10-26 15:52               ` Octavian Purdila
2009-10-26 16:55                 ` Stephen Hemminger [this message]
2009-10-26 17:45                   ` Stephen Hemminger
2009-10-27  1:24               ` David Miller
2009-10-27  1:40                 ` Eric Dumazet
2009-10-26  6:30   ` Stephen Hemminger
2009-10-26  7:48     ` Eric Dumazet
2009-10-26  4:43 ` Stephen Hemminger
2009-10-26 22:36   ` [PATCH] dcache: better name hash function Stephen Hemminger <shemminger@vyatta.com>, Al Viro
2009-10-27  2:45     ` Eric Dumazet
2009-10-27  3:53       ` Stephen Hemminger
2009-10-27 16:38       ` Rick Jones

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20091026095516.02f1cb49@nehalam \
    --to=shemminger@vyatta.com \
    --cc=eric.dumazet@gmail.com \
    --cc=hagen@jauu.net \
    --cc=krkumar2@in.ibm.com \
    --cc=netdev@vger.kernel.org \
    --cc=opurdila@ixiacom.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).