[wsf-c-dev] Axis2/C performance resullts with fine tuned Apache httpd
Rajika Kumarasiri
rajika at wso2.com
Mon Jul 7 22:23:52 PDT 2008
On Tue, 2008-07-08 at 10:47 +0530, Samisa Abeysinghe wrote:
> Rajika Kumarasiri wrote:
> > On Tue, 2008-07-08 at 08:10 +0530, Samisa Abeysinghe wrote:
> >
> >> What is the fastest Web server around? I suppose it is not httpd.
> >>
> > Nope, Yaws ?
> > http://yaws.hyber.org/
> >
> > Apache vs. Yaws,
> > http://www.sics.se/~joe/apachevsyaws.html
> >
> > BTW, yaws was written in Erlang.
> >
>
> BTW, if that is the case, then that does not work for us, does it? Can
> we write an extension?
I don't know weather they have a native C API, will update soon.
-Rajika
>
> Samisa...
>
> > -Rajika
> >
> >> What if we write an Axis2/C module for that fastest Web server?
> >>
> >> Samisa...
> >>
> >>
> >> Rajika Kumarasiri wrote:
> >>
> >>> Samisa Abeysinghe wrote:
> >>>
> >>>> Rajika Kumarasiri wrote:
> >>>>
> >>>>> Samisa Abeysinghe wrote:
> >>>>>
> >>>>>> Samisa Abeysinghe wrote:
> >>>>>>
> >>>>>>> What were the kernel tuning that were done?
> >>>>>>>
> >>>>>>> Also, only 16,000 tps with httpd for pure html? Given the
> >>>>>>> processing power of our test environment, this should not it be
> >>>>>>> far better than this?
> >>>>>>>
> >>>>>> http://httpd.apache.org/docs/2.0/misc/perf-tuning.html
> >>>>>> http://www.devside.net/articles/apache-performance-tuning
> >>>>>>
> >>>>> Well, I followed this and the guidance available here.
> >>>>> http://www.stdlib.net/~colmmacc/Apachecon-EU2005/scaling-apache-handout.pdf
> >>>>>
> >>>>>
> >>>> There are some additional info in the above links. Like removing
> >>>> unused modules.
> >>>>
> >>>> My gut feel is that, normal httpd could perform way beyond 16k TPS. I
> >>>> could be wrong. But we need to research this.
> >>>>
> >>> I 'll further look into this.
> >>>
> >>> -Rajika
> >>>
> >>>> Thanks,
> >>>> Samisa...
> >>>>
> >>>>
> >>>>> Any way, thanks for the information.
> >>>>>
> >>>>> -Rajika
> >>>>>
> >>>>>> Thanks,
> >>>>>> Samisa...
> >>>>>>
> >>>>>>> Thanks,
> >>>>>>> Samisa...
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>> Rajika Kumarasiri wrote:
> >>>>>>>
> >>>>>>>> hello every body,
> >>>>>>>> I did a performance test for Axis2/C using httpd module,
> >>>>>>>> mod_axis2. The
> >>>>>>>> test was carried out after fine tuning(for virtual memory, file
> >>>>>>>> system,
> >>>>>>>> network etc..) the Apache web server.
> >>>>>>>> Before fine tuning Apache, it was able to serve around 13,000
> >>>>>>>> requests
> >>>>>>>> per second (I checked this for static content of size 44 bytes,
> >>>>>>>> using
> >>>>>>>> index.html as a sample) and after the tuning it was able to serve
> >>>>>>>> around
> >>>>>>>> 16,000 requests per second.
> >>>>>>>> The test was done in the same manner as the previous one using the
> >>>>>>>> Guththila as the parser under same hardware resources.
> >>>>>>>>
> >>>>>>>> Results are as follows.(in requests per second)
> >>>>>>>>
> >>>>>>>> small data set (with 1 element)
> >>>>>>>> ==============
> >>>>>>>> echoVoid(269 bytes) - 12293
> >>>>>>>> echoDoubles(298 bytes) - 11666
> >>>>>>>> echoInts(290 bytes) - 12095
> >>>>>>>> echoStrings(297 bytes) - 12,033
> >>>>>>>> echoSimpleEvents(418 bytes) - 11,440
> >>>>>>>> echoMeshInterfaceObjects(379 bytes) - 11861
> >>>>>>>>
> >>>>>>>> large data set (with 100 elements) ==============
> >>>>>>>> echoDoubles(2.9 KB) - 5676
> >>>>>>>> echoInts(2.7 KB) - 6816.83
> >>>>>>>> echoStrings(2.8 KB) - 6984.28
> >>>>>>>> echoSimpleEvents(15 KB) - 1756.59
> >>>>>>>> echoMeshInterfaceObjects(9.6 KB) - 2204.84
> >>>>>>>>
> >>>>>>>> Conclusion:
> >>>>>>>> According to the results there is around 700- 1500 requests per
> >>>>>>>> seconds
> >>>>>>>> improvement for small data and for large data the amounts remain the
> >>>>>>>> same as the previous.
> >>>>>>>>
> >>>>>>>> -Rajika
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> _______________________________________________
> >>>>>>>> Wsf-c-dev mailing list
> >>>>>>>> Wsf-c-dev at wso2.org
> >>>>>>>> http://wso2.org/cgi-bin/mailman/listinfo/wsf-c-dev
> >>>>>>>> ------------------------------------------------------------------------
> >>>>>>>>
> >>>>>>>>
> >>>>>>>>
> >>>>>>>> No virus found in this incoming message.
> >>>>>>>> Checked by AVG. Version: 8.0.100 / Virus Database: 270.0.0/1489 -
> >>>>>>>> Release Date: 6/7/2008 11:17 AM
> >>>>>>>>
> >>>>>>>>
> >>>>>>> ------------------------------------------------------------------------
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>>> No virus found in this incoming message.
> >>>>>>> Checked by AVG. Version: 8.0.100 / Virus Database: 270.0.0/1489 -
> >>>>>>> Release Date: 6/7/2008 11:17 AM
> >>>>>>>
> >>>>>>>
> >>>>>>
> >>>>> _______________________________________________
> >>>>> Wsf-c-dev mailing list
> >>>>> Wsf-c-dev at wso2.org
> >>>>> http://wso2.org/cgi-bin/mailman/listinfo/wsf-c-dev
> >>>>> ------------------------------------------------------------------------
> >>>>>
> >>>>>
> >>>>>
> >>>>> No virus found in this incoming message.
> >>>>> Checked by AVG. Version: 8.0.100 / Virus Database: 270.0.0/1489 -
> >>>>> Release Date: 6/7/2008 11:17 AM
> >>>>>
> >>>>>
> >>>>
> >>> _______________________________________________
> >>> Wsf-c-dev mailing list
> >>> Wsf-c-dev at wso2.org
> >>> http://wso2.org/cgi-bin/mailman/listinfo/wsf-c-dev
> >>> ------------------------------------------------------------------------
> >>>
> >>>
> >>> No virus found in this incoming message.
> >>> Checked by AVG.
> >>> Version: 8.0.100 / Virus Database: 270.0.0/1489 - Release Date: 6/7/2008 11:17 AM
> >>>
> >>>
> >>
> >
> >
> > _______________________________________________
> > Wsf-c-dev mailing list
> > Wsf-c-dev at wso2.org
> > http://wso2.org/cgi-bin/mailman/listinfo/wsf-c-dev
> >
> > No virus found in this incoming message.
> > Checked by AVG - http://www.avg.com
> > Version: 8.0.138 / Virus Database: 270.4.6/1538 - Release Date: 7/7/2008 7:40 AM
> >
> >
> >
> >
>
>
More information about the Wsf-c-dev
mailing list