<p class="MsoNormal">Thanks a lot for all your help and efforts.</p><p class="MsoNormal"><br></p>
<p class="MsoNormal">Any suggestion what are potential configurations I may have
to do to increase number of open file and socket on windows to make my test
scalable to 15K?</p><p class="MsoNormal"><br></p><p class="MsoNormal">I have tried searching OS file handler and socket related configuration on windows, had no success so far.</p><p class="MsoNormal"><br></p><p class="MsoNormal">
regards,</p><p class="MsoNormal">Jaymin</p><p class="MsoNormal"><br></p><br><div class="gmail_quote">On Thu, Aug 11, 2011 at 4:30 PM, Jamie McCrindle <span dir="ltr"><<a href="mailto:jamiemccrindle@gmail.com">jamiemccrindle@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">It's worth saying that at 20000 requests and 200 threads, a default<br>
Mac OS X 10.6 installation falls apart as it runs out of open files,<br>
sockets get stuck in TIME_WAIT etc. so there is definitely an OS<br>
tuning component to running a test like this.<br>
<div><div></div><div class="h5"><br>
On Thu, Aug 11, 2011 at 9:49 AM, Jamie McCrindle<br>
<<a href="mailto:jamiemccrindle@gmail.com">jamiemccrindle@gmail.com</a>> wrote:<br>
> Hi Jaymin,<br>
><br>
> Here are the results on my machine. It's Mac OS X 10.6, MBP 2.66 GHZ<br>
> i7. I had to drop the numbers to 10000 requests and 100 threads with<br>
> the default Mac config. This is with the snoop HttpServer run as<br>
> follows:<br>
><br>
> mvn -Dexec.mainClass=org.jboss.netty.example.http.snoop.HttpServerDmaven.test.skip=true<br>
> compile -Dskip-enforce=true exec:java<br>
><br>
> It may be worth dropping the number of requests you're testing with as<br>
> well as the concurrency.<br>
><br>
> headline figure is: Requests per second: 13070.43 [#/sec] (mean)<br>
><br>
> ab -n 10000 -c 100 "<a href="http://localhost:8080/" target="_blank">http://localhost:8080/</a>"<br>
> This is ApacheBench, Version 2.3 <$Revision: 655654 $><br>
> Copyright 1996 Adam Twiss, Zeus Technology Ltd, <a href="http://www.zeustech.net/" target="_blank">http://www.zeustech.net/</a><br>
> Licensed to The Apache Software Foundation, <a href="http://www.apache.org/" target="_blank">http://www.apache.org/</a><br>
><br>
> Benchmarking localhost (be patient)<br>
> Completed 1000 requests<br>
> Completed 2000 requests<br>
> Completed 3000 requests<br>
> Completed 4000 requests<br>
> Completed 5000 requests<br>
> Completed 6000 requests<br>
> Completed 7000 requests<br>
> Completed 8000 requests<br>
> Completed 9000 requests<br>
> Completed 10000 requests<br>
> Finished 10000 requests<br>
><br>
><br>
> Server Software:<br>
> Server Hostname: localhost<br>
> Server Port: 8080<br>
><br>
> Document Path: /<br>
> Document Length: 230 bytes<br>
><br>
> Concurrency Level: 100<br>
> Time taken for tests: 0.765 seconds<br>
> Complete requests: 10000<br>
> Failed requests: 5<br>
> (Connect: 5, Receive: 0, Length: 0, Exceptions: 0)<br>
> Write errors: 0<br>
> Total transferred: 2901160 bytes<br>
> HTML transferred: 2300920 bytes<br>
> Requests per second: 13070.43 [#/sec] (mean)<br>
> Time per request: 7.651 [ms] (mean)<br>
> Time per request: 0.077 [ms] (mean, across all concurrent requests)<br>
> Transfer rate: 3703.07 [Kbytes/sec] received<br>
><br>
> Connection Times (ms)<br>
> min mean[+/-sd] median max<br>
> Connect: 0 1 0.6 1 7<br>
> Processing: 0 3 2.2 3 14<br>
> Waiting: 0 3 2.0 2 13<br>
> Total: 1 4 2.2 4 14<br>
><br>
> Percentage of the requests served within a certain time (ms)<br>
> 50% 4<br>
> 66% 4<br>
> 75% 5<br>
> 80% 5<br>
> 90% 7<br>
> 95% 9<br>
> 98% 11<br>
> 99% 12<br>
> 100% 14 (longest request)<br>
><br>
> On Thu, Aug 11, 2011 at 9:06 AM, Jaymin Shah <<a href="mailto:sjaymin@gmail.com">sjaymin@gmail.com</a>> wrote:<br>
>> I had tried it never got throughput more then 3K.<br>
>> So question is can Netty be able to achieve mentioned throughput more then<br>
>> 15K as per <a href="http://www.olympum.com/java/quick-benchmark-java-nodejs/" target="_blank">http://www.olympum.com/java/quick-benchmark-java-nodejs/</a> ?<br>
>><br>
>><br>
>> On Thu, Aug 11, 2011 at 1:28 PM, Mailing List SVR <<a href="mailto:lists@svrinformatica.it">lists@svrinformatica.it</a>><br>
>> wrote:<br>
>>><br>
>>> Il 11/08/2011 09:34, Jaymin Shah ha scritto:<br>
>>><br>
>>> Please find below command I had used and its output.<br>
>>> Command#<br>
>>> ab -n 20000 -c 200 <a href="http://localhost:8080/" target="_blank">http://localhost:8080/</a><br>
>>><br>
>>> please try with different concurrency for example -c 5, -c 10 and so on<br>
>>> and see if the result change,<br>
>>><br>
>>> Nicola<br>
>>><br>
>>><br>
>>> Output#<br>
>>> This is ApacheBench, Version 2.3 <$Revision: 655654 $><br>
>>> Copyright 1996 Adam Twiss, Zeus Technology Ltd, <a href="http://www.zeustech.net/" target="_blank">http://www.zeustech.net/</a><br>
>>> Licensed to The Apache Software Foundation, <a href="http://www.apache.org/" target="_blank">http://www.apache.org/</a><br>
>>> Benchmarking localhost (be patient)<br>
>>> Completed 2000 requests<br>
>>> Completed 4000 requests<br>
>>> Completed 6000 requests<br>
>>> Completed 8000 requests<br>
>>> Completed 10000 requests<br>
>>> Completed 12000 requests<br>
>>> Completed 14000 requests<br>
>>> Completed 16000 requests<br>
>>> Completed 18000 requests<br>
>>> Completed 20000 requests<br>
>>> Finished 20000 requests<br>
>>><br>
>>> Server Software:<br>
>>> Server Hostname: localhost<br>
>>> Server Port: 8080<br>
>>> Document Path: /<br>
>>> Document Length: 230 bytes<br>
>>> Concurrency Level: 200<br>
>>> Time taken for tests: 8.438 seconds<br>
>>> Complete requests: 20000<br>
>>> Failed requests: 0<br>
>>> Write errors: 0<br>
>>> Total transferred: 5800000 bytes<br>
>>> HTML transferred: 4600000 bytes<br>
>>> Requests per second: 2370.09 [#/sec] (mean)<br>
>>> Time per request: 84.385 [ms] (mean)<br>
>>> Time per request: 0.422 [ms] (mean, across all concurrent requests)<br>
>>> Transfer rate: 671.22 [Kbytes/sec] received<br>
>>> Connection Times (ms)<br>
>>> min mean[+/-sd] median max<br>
>>> Connect: 0 0 6.1 0 507<br>
>>> Processing: 5 83 97.4 63 1099<br>
>>> Waiting: 3 52 76.5 41 598<br>
>>> Total: 5 83 97.7 64 1099<br>
>>> Percentage of the requests served within a certain time (ms)<br>
>>> 50% 64<br>
>>> 66% 70<br>
>>> 75% 77<br>
>>> 80% 80<br>
>>> 90% 101<br>
>>> 95% 116<br>
>>> 98% 587<br>
>>> 99% 597<br>
>>> 100% 1099 (longest request)<br>
>>><br>
>>> On Thu, Aug 11, 2011 at 9:36 AM, News Aanad <<a href="mailto:news.anand11@gmail.com">news.anand11@gmail.com</a>><br>
>>> wrote:<br>
>>>><br>
>>>><br>
>>>> ---------- Forwarded message ----------<br>
>>>> From: Jamie McCrindle <<a href="mailto:jamiemccrindle@gmail.com">jamiemccrindle@gmail.com</a>><br>
>>>> Date: Wed, Aug 10, 2011 at 11:53 AM<br>
>>>> Subject: Re: netty based http server performance 15K/sec<br>
>>>> To: Netty Users <<a href="mailto:netty-users@lists.jboss.org">netty-users@lists.jboss.org</a>><br>
>>>><br>
>>>><br>
>>>> Can you send the apache bench command line you used along with the<br>
>>>> output?<br>
>>>><br>
>>>> On Aug 10, 2011 8:09 AM, "Jaymin Shah" <<a href="mailto:sjaymin@gmail.com">sjaymin@gmail.com</a>> wrote:<br>
>>>> > Hi,<br>
>>>> ><br>
>>>> > It was claimed that netty based http server can handle *15693.29<br>
>>>> > [#/sec]<br>
>>>> > request* ref# <a href="http://www.olympum.com/java/quick-benchmark-java-nodejs/" target="_blank">http://www.olympum.com/java/quick-benchmark-java-nodejs/</a><br>
>>>> ><br>
>>>> > I have tried using org.jboss.netty.example.http.snoop but never got<br>
>>>> > performance beyond* 2285.58 [#/sec] (mean)*. I am using Windows 7 -<br>
>>>> > 64-bit<br>
>>>> > operating system with 4GB RAM and Intel core i5 with 2.40GHz<br>
>>>> ><br>
>>>> > Can someone please share their experience how I can achieve mentioned<br>
>>>> > performance.<br>
>>>> ><br>
>>>> > regards,<br>
>>>> > Jaymin<br>
>>>><br>
>>>> _______________________________________________<br>
>>>> netty-users mailing list<br>
>>>> <a href="mailto:netty-users@lists.jboss.org">netty-users@lists.jboss.org</a><br>
>>>> <a href="https://lists.jboss.org/mailman/listinfo/netty-users" target="_blank">https://lists.jboss.org/mailman/listinfo/netty-users</a><br>
>>>><br>
>>><br>
>>><br>
>>><br>
>>> _______________________________________________<br>
>>> netty-users mailing list<br>
>>> <a href="mailto:netty-users@lists.jboss.org">netty-users@lists.jboss.org</a><br>
>>> <a href="https://lists.jboss.org/mailman/listinfo/netty-users" target="_blank">https://lists.jboss.org/mailman/listinfo/netty-users</a><br>
>>><br>
>>> _______________________________________________<br>
>>> netty-users mailing list<br>
>>> <a href="mailto:netty-users@lists.jboss.org">netty-users@lists.jboss.org</a><br>
>>> <a href="https://lists.jboss.org/mailman/listinfo/netty-users" target="_blank">https://lists.jboss.org/mailman/listinfo/netty-users</a><br>
>><br>
>><br>
>> _______________________________________________<br>
>> netty-users mailing list<br>
>> <a href="mailto:netty-users@lists.jboss.org">netty-users@lists.jboss.org</a><br>
>> <a href="https://lists.jboss.org/mailman/listinfo/netty-users" target="_blank">https://lists.jboss.org/mailman/listinfo/netty-users</a><br>
>><br>
><br>
<br>
_______________________________________________<br>
netty-users mailing list<br>
<a href="mailto:netty-users@lists.jboss.org">netty-users@lists.jboss.org</a><br>
<a href="https://lists.jboss.org/mailman/listinfo/netty-users" target="_blank">https://lists.jboss.org/mailman/listinfo/netty-users</a><br>
</div></div></blockquote></div><br>