Re: Sync Rep and shutdown Re: Sync Rep v19 - Mailing list pgsql-hackers

On 2011-03-21 23:58, Yeb Havinga wrote: <blockquote
cite="mid:AANLkTinYRqDU=zkJSPAy+f2BiYhNCX_NomtDRTBWNvA4@mail.gmail.com"type="cite"><div class="gmail_quote">On Mon, Mar
21,2011 at 7:51 PM, Yeb Havinga <span dir="ltr"><<a href="mailto:yebhavinga@gmail.com"
moz-do-not-send="true">yebhavinga@gmail.com</a>></span>wrote:<br /><blockquote class="gmail_quote" style="margin:
0pt0pt 0pt         0.8ex; border-left: 1px solid rgb(204, 204, 204);         padding-left: 1ex;"><div class="im">On
2011-03-2118:04, Robert Haas wrote:<br /><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt             0.8ex;
border-left:1px solid rgb(204, 204, 204);             padding-left: 1ex;"> On Mon, Mar 21, 2011 at 12:29 PM, Yeb
Havinga<<ahref="mailto:yebhavinga@gmail.com" moz-do-not-send="true" target="_blank">yebhavinga@gmail.com</a>>
 wrote:<br/><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt               0.8ex; border-left: 1px solid
rgb(204,204, 204);               padding-left: 1ex;"> pgbench -i -s 50 test<br /> Two runs of "pgbench -c 10 -M
prepared-T 600 test" with 1 sync standby -<br /> server configs etc were mailed upthread.<br /><br /><blockquote
class="gmail_quote"style="margin: 0pt 0pt                 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204);
       padding-left: 1ex;"> - performance as of commit e148443ddd95cd29edf4cc1de6188eb9cee029c5<br /></blockquote> 1158
and1306 (avg 1232)<br /><blockquote class="gmail_quote" style="margin: 0pt 0pt                 0pt 0.8ex; border-left:
1pxsolid rgb(204, 204, 204);                 padding-left: 1ex;"> - performance as of current git master<br
/></blockquote>1181 and 1280 (avg 1230,5)<br /><blockquote class="gmail_quote" style="margin: 0pt 0pt
0pt0.8ex; border-left: 1px solid rgb(204, 204, 204);                 padding-left: 1ex;"> - performance as of current
gitmaster with<br /> sync-standbys-defined-rearrangement applied<br /></blockquote> 1152 and 1269 (avg 1210,5)<br
/></blockquote></blockquote><br/></div><br /> IMO what these tests have shown is that there is no 20% performance
differencebetween the different versions. To determine if there are differences, n should be a lot higher, or perhaps a
singleone with a very large duration.</blockquote><br /> pgbench -T 3600:<br /><br />
sync-standbys-defined-rearrangement1270 tps<br /> current git master 1306 tps<br /></div></blockquote><br /> Result of
pgbench-T 30000<br /> sync-standbys-defined-rearrangement 1267 tps<br /> current (or few days old) git master 1326
tps<br/><br /> So the patch eats 4,5% from git master's syncrep performance in my setup. Don't know how to measure it
betterthan that.<br /><br /><pre class="moz-signature" cols="72">-- 
 
Yeb Havinga
<a class="moz-txt-link-freetext" href="http://www.mgrid.net/">http://www.mgrid.net/</a>
Mastering Medical Data</pre>