I should clarify that this AMD machine is a test server, and has CoolnQuiet turned on in<br>the BIOS. So, it reduces its frequency to 1GHz when idle, and goes up to 1.8 when busy.<br>I can't remember if the disk has power saving too or not, but that can account for the
<br>variance. <br><br>This is why the last benchmark did 1000 requests in one row,<br><br>The other factor is MySQL query cache, but again, that would be off at most for the first<br>request, and on for the 999 following it, so it should not be a factor in the last test.
<br><br>Can someone else chip in with other benchmarks to validate my results?<br><br><div><span class="gmail_quote">On 3/5/07, <b class="gmail_sendername">Robert Douglass</b> <<a href="mailto:rob@robshouse.net">rob@robshouse.net
</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">I wonder right off why HEAD's times are so much more level and<br>
predictable (smaller standard deviation) than 5's? Especially since my<br>gut feeling is that 5 is much more predictable and regular than 4.7.<br><br>Congrats to chx and company for this work... it seems that Drupal 6 will
<br>indeed be faster than 5 by virtue of the menu rewrite.<br><br><br>Khalid Baheyeldin wrote:<br>> Here is a better set of data:<br>><br>> HEAD<br>> ab2 -n20 <a href="http://head.example.com/|">http://head.example.com/|
</a> <<a href="http://head.example.com/%7C">http://head.example.com/%7C</a>> egrep<br>> "^Req|^Time"<br>><br>> Time taken for tests: 1.16209 seconds<br>> Requests per second: 19.68 [#/sec] (mean)
<br>> Time per request: 50.810 [ms] (mean)<br>> Time per request: 50.810 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 1.11599 seconds<br>> Requests per second:
19.77 [#/sec] (mean)<br>> Time per request: 50.580 [ms] (mean)<br>> Time per request: 50.580 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 1.10162 seconds<br>> Requests per second:
19.80 [#/sec] (mean)<br>> Time per request: 50.508 [ms] (mean)<br>> Time per request: 50.508 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 1.14755 seconds<br>> Requests per second:
19.71 [#/sec] (mean)<br>> Time per request: 50.738 [ms] (mean)<br>> Time per request: 50.738 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 1.11669 seconds<br>> Requests per second:
19.77 [#/sec] (mean)<br>> Time per request: 50.583 [ms] (mean)<br>> Time per request: 50.583 [ms] (mean, across all concurrent requests)<br>><br>> 5<br>> ab2 -n20 <a href="http://5.example.com/|">
http://5.example.com/|</a> <<a href="http://5.example.com/%7C">http://5.example.com/%7C</a>> egrep<br>> "^Req|^Time"<br>> Time taken for tests: 2.672659 seconds<br>> Requests per second: 7.48
[#/sec] (mean)<br>> Time per request: 133.633 [ms] (mean)<br>> Time per request: 133.633 [ms] (mean, across all concurrent<br>> requests)<br>><br>> Time taken for tests: 1.87295 seconds<br>> Requests per second:
18.39 [#/sec] (mean)<br>> Time per request: 54.365 [ms] (mean)<br>> Time per request: 54.365 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 1.55900 seconds<br>> Requests per second:
18.94 [#/sec] (mean)<br>> Time per request: 52.795 [ms] (mean)<br>> Time per request: 52.795 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 0.887638 seconds<br>> Requests per second:
22.53 [#/sec] (mean)<br>> Time per request: 44.382 [ms] (mean)<br>> Time per request: 44.382 [ms] (mean, across all concurrent requests)<br>><br>> Time taken for tests: 1.90529 seconds<br>> Requests per second:
18.34 [#/sec] (mean)<br>> Time per request: 54.526 [ms] (mean)<br>> Time per request: 54.526 [ms] (mean, across all concurrent requests)<br>><br>><br>> And here are two tests with 1000 requests each:
<br>><br>> ab2 -n1000 <a href="http://head.example.com/|">http://head.example.com/|</a> <<a href="http://head.example.com/%7C">http://head.example.com/%7C</a>><br>> egrep "^Req|^Time"<br>> Completed 100 requests
<br>> Completed 200 requests<br>> Completed 300 requests<br>> Completed 400 requests<br>> Completed 500 requests<br>> Completed 600 requests<br>> Completed 700 requests<br>> Completed 800 requests<br>
> Completed 900 requests<br>> Finished 1000 requests<br>> Time taken for tests: 33.209431 seconds<br>> Requests per second: 30.11 [#/sec] (mean)<br>> Time per request: 33.209 [ms] (mean)<br>> Time per request:
33.209 [ms] (mean, across all concurrent requests)<br>><br>> ab2 -n1000 <a href="http://5.example.com/|">http://5.example.com/|</a> <<a href="http://5.example.com/%7C">http://5.example.com/%7C</a>> egrep<br>> "^Req|^Time"
<br>> Completed 100 requests<br>> Completed 200 requests<br>> Completed 300 requests<br>> Completed 400 requests<br>> Completed 500 requests<br>> Completed 600 requests<br>> Completed 700 requests<br>
> Completed 800 requests<br>> Completed 900 requests<br>> Finished 1000 requests<br>> Time taken for tests: 37.155010 seconds<br>> Requests per second: 26.91 [#/sec] (mean)<br>> Time per request:
37.155 [ms] (mean)<br>> Time per request: 37.155 [ms] (mean, across all concurrent requests)<br>><br>> On 3/4/07, *Khalid Baheyeldin* <<a href="mailto:kb@2bits.com">kb@2bits.com</a> <mailto:<a href="mailto:kb@2bits.com">
kb@2bits.com</a>>><br>> wrote:<br>><br>> Here are my results.<br>><br>> Drupal 5, is today's checkout, HEAD is also today's checkout.<br>><br>> Used ab2 -n20 <a href="http://5.example.com">
http://5.example.com</a><br>> HEAD<br>> Requests per second: 19.78 [#/sec] (mean)<br>> Requests per second: 23.93 [#/sec] (mean)<br>> Requests per second: 19.81 [#/sec] (mean)<br>> Requests per second:
19.85 [#/sec] (mean)<br>><br>> Used ab2 -n20 <a href="http://head.example.com">http://head.example.com</a> <<a href="http://head.example.com">http://head.example.com</a>><br>> Drupal 5<br>> Requests per second:
18.58 [#/sec] (mean)<br>> Requests per second: 21.73 [#/sec] (mean)<br>> Requests per second: 18.35 [#/sec] (mean)<br>> Requests per second: 28.12 [#/sec] (mean)<br>><br>> Server is an AMD 64 3000+ with a SATA 7200 Disk with 1GB RAM
<br>> Ubuntu Edgy 6.10, Apache 2.0.55, MySQL 5.0.24, PHP 5.1.6<br>><br>><br>> On 3/3/07, *Karoly Negyesi* < <a href="mailto:karoly@negyesi.net">karoly@negyesi.net</a><br>> <mailto:<a href="mailto:karoly@negyesi.net">
karoly@negyesi.net</a>>> wrote:<br>><br>> It is there. Please try enabling all core modules in Drupal 5<br>> and bench the front page as anon. Then try the same in Drupal<br>> 6-dev.I
am really interested in results.<br>><br>> > In particular, we need benchmarking help, soon I will post a<br>> simpler version which does a permission check for every menu<br>> item. I wonder how bad that is...
<br>> ><br>><br>><br>><br>><br>> --<br>> <a href="http://2bits.com">2bits.com</a> <<a href="http://2bits.com">http://2bits.com</a>><br>> <a href="http://2bits.com">http://2bits.com
</a><br>> Drupal development, customization and consulting.<br>><br>><br>><br>><br>> --<br>> <a href="http://2bits.com">2bits.com</a> <<a href="http://2bits.com">http://2bits.com</a>><br>>
<a href="http://2bits.com">http://2bits.com</a><br>> Drupal development, customization and consulting.<br><br><br>--<br>* * * * *<br>Lullabot's First Ever Advanced Workshops Are Here!<br>Drupal API & Module Building - Advanced Drupal Themeing
<br>April 9th-13th - Providence, RI<br>Early Bird Discounts Available Now<br><a href="http://www.lullabot.com/training">http://www.lullabot.com/training</a><br>* * * * *<br><br></blockquote></div><br><br clear="all"><br>--
<br><a href="http://2bits.com">2bits.com</a><br><a href="http://2bits.com">http://2bits.com</a><br>Drupal development, customization and consulting.