<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Levente,<div><br><div><div>On 12 Oct 2010, at 15:42, Levente Uzonyi wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite"><span class="Apple-style-span" style="border-collapse: separate; font-family: Helvetica; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0px; font-size: medium; ">That's really interesting. Our benchmark showed almost no speedup at all for Kom + Seaside3 with Cog. What was your benchmark like?<br></span></blockquote></div><br><div>Using the latest Seaside 3 nightly build (from <a href="http://hudson.lukas-renggli.ch/">http://hudson.lukas-renggli.ch/</a>), using MC to get my DWBench benchmark (essentially a dynamically generated 25 by 25 products table of about 8Kb) from the ADayAtTheBeach project on SqueakSource, I get the following results:</div><div><br></div><div>Old VM (Squeak 4.2.5beta1U on MacOSX)</div><div><br></div><div><font class="Apple-style-span" face="Courier">[sven@voyager:~]$ ab -d -n 64 -c 4 <a href="http://127.0.0.1:8080/DW-Bench">http://127.0.0.1:8080/DW-Bench</a><br>This is ApacheBench, Version 2.3 <$Revision: 655654 $><br>Copyright 1996 Adam Twiss, Zeus Technology Ltd, <a href="http://www.zeustech.net/">http://www.zeustech.net/</a><br>Licensed to The Apache Software Foundation, <a href="http://www.apache.org/">http://www.apache.org/</a><br><br>Benchmarking 127.0.0.1 (be patient).....done<br><br><br>Server Software: KomHttpServer/7.1.3<br>Server Hostname: 127.0.0.1<br>Server Port: 8080<br><br>Document Path: /DW-Bench<br>Document Length: 8899 bytes<br><br>Concurrency Level: 4<br>Time taken for tests: 0.912 seconds<br>Complete requests: 64<br>Failed requests: 0<br>Write errors: 0<br>Total transferred: 580736 bytes<br>HTML transferred: 569536 bytes<br>Requests per second: 70.14 [#/sec] (mean)<br>Time per request: 57.030 [ms] (mean)<br>Time per request: 14.257 [ms] (mean, across all concurrent requests)<br>Transfer rate: 621.53 [Kbytes/sec] received<br><br>Connection Times (ms)<br> min mean[+/-sd] median max<br>Connect: 0 0 1.3 0 10<br>Processing: 15 56 15.3 55 109<br>Waiting: 15 55 15.3 55 109<br>Total: 15 56 15.3 55 109<br></font><br></div><div>New VM (Squeak 5.8b12 on MacOSX)</div><div><br></div><font class="Apple-style-span" face="Courier">[sven@voyager:~]$ ab -d -n 64 -c 4 <a href="http://127.0.0.1:8080/DW-Bench">http://127.0.0.1:8080/DW-Bench</a><br>This is ApacheBench, Version 2.3 <$Revision: 655654 $><br>Copyright 1996 Adam Twiss, Zeus Technology Ltd, <a href="http://www.zeustech.net/">http://www.zeustech.net/</a><br>Licensed to The Apache Software Foundation, <a href="http://www.apache.org/">http://www.apache.org/</a><br><br>Benchmarking 127.0.0.1 (be patient).....done<br><br><br>Server Software: KomHttpServer/7.1.3<br>Server Hostname: 127.0.0.1<br>Server Port: 8080<br><br>Document Path: /DW-Bench<br>Document Length: 8899 bytes<br><br>Concurrency Level: 4<br>Time taken for tests: 0.301 seconds<br>Complete requests: 64<br>Failed requests: 0<br>Write errors: 0<br>Total transferred: 580736 bytes<br>HTML transferred: 569536 bytes<br>Requests per second: 212.74 [#/sec] (mean)<br>Time per request: 18.802 [ms] (mean)<br>Time per request: 4.701 [ms] (mean, across all concurrent requests)<br>Transfer rate: 1885.14 [Kbytes/sec] received<br><br>Connection Times (ms)<br> min mean[+/-sd] median max<br>Connect: 0 0 0.1 0 1<br>Processing: 4 18 2.6 19 21<br>Waiting: 4 18 2.6 19 21<br>Total: 4 18 2.6 19 21<br></font><br><div><br></div><div>Now, this is actually a (concurrent) session creation benchmark if you think about it. ab and almost all other benchmark tools have a really hard time to track seaside sessions. But the request handling is real, as is the actual speedup.</div><div><br></div><div>Sven</div><div><br></div><div><br><div><br></div><div> </div></div></div></body></html>