jmeter-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Philippe Mouawad <philippe.moua...@gmail.com>
Subject Re: PERFORMANCE => Changing JMeter defaults to ensure better performances by default
Date Tue, 08 Jan 2013 15:26:21 GMT
Hello,
The groovy-all.jar in in *embeddable* folder of the groovy-binary-2.0.6.zip
file.

Once you add it in jmeter/lib folder, restart jmeter and groovy will appear.
I will do some autopromotion :-) , see this for what will be available in
JMeter 2.9 (part of it is in 2.8):

   -
   http://www.ubik-ingenierie.com/blog/jmeter_control_percentage_of_sampler/


If you want to test, you can get nightly build  (but read warning and
install instructions):

   - http://jmeter.apache.org/nightly.html

Regards

Philippe

http://www.ubik-ingenierie.com/blog/
On Tue, Jan 8, 2013 at 3:49 PM, Keith Young <youngka@gmail.com> wrote:

> >> Encourage JSR223 Samplers + Groovy  + Caching instead of Beanshell
>
> Can someone point me to some documentation that outlines this better?  If I
> include a JSR223 sampler there is no option for Groovy.  The wiki link in
> this message thread says to include groovy-VERSION-all.jar in the
> jmeter/bin directory, but a download of Groovy 2.0.6 doesn't have that jar.
> Is the groovy-2.0.6.jar enough?
>
> What caching is the above point referring to?
>
> Sorry for these questions on the dev list, but that's where this thread
> is... :)
>
> Cheers,
> Keith.
>
>
> On 27 December 2012 05:06, Milamber <milamber@apache.org> wrote:
>
> >
> >
> > Le 24/12/2012 13:20, Philippe Mouawad a ecrit :
> >
> >  Hello,
> >>
> >> I am kind of annoyed of reading articles, blogs that say JMeter cannot
> >> perform high Load Tests, consumes lot of memory, generates OutOfMemory
> ...
> >>
> >> This has become a kind of "Urban Legend" partly due:
> >> - to issues that have been fixed for a while now
> >> - and partly In my opinion to some default configuration parameters that
> >> lead to these issues
> >>
> >> In my opinion, we should:
> >>
> >> 1) change these defaults to avoid new comers, beginners fall into all
> >> these
> >> traps and others check they are using it well:
> >>
> >>     - Save Service using XML output =>  Change to CSV
> >>
> >
> > Yes, CSV seems a better default type to the results file.
> >
> >
> >      - Distributed Mode that uses the Standard which is far from being
> the
> >>     best performing Sample Sender =>  Change to Batch or StrippedBatch
> >>
> >
> > Why Batch ou StrippedBatch ? Why not Asynch or Diskstore?
> >
> >
> >
> >> 2) Add warnings on GUIs of all elements that are more suited during
> >> Scripting than during Load Test :
> >>
> >>     - View Result Tree (I keep seeing people use this element during
> High
> >>     Load Test ! )
> >>     - View Results in Table
> >>     - Graph Results
> >>     - ...
> >>
> >
> > JMeter can be used for functional tests. Warnings on GUI for the 'heavy'
> > elements can be unpleasant.
> >
> > Perhaps add a menu/button option "Check script for load test" which
> > display some warning (in a box or another way) when heavy elements are
> > found in the tree.
> > This option can be enabled by a property to checked before every load
> tests
> >
> >
> >  3) Add a popup warning when Start and Remote Start are clicked from GUI
> to
> >> encourage NON GUI mode use (we could add a checkbox Remind Me later
> which
> >> could be unchecked to avoid it again, but at least user would know about
> >> it).
> >>
> >
> > See comment in 2)
> >
> >
> >
> >
> >  4) Finally use some kind of visual indicator (RED background) on some
> >> options that have high impact on performance:
> >>
> >>     - Javascript as scripting language
> >>     - Body (unescaped) in Regular Expression Extractor (*this one is a
> >> real
> >>     performance killer !*)
> >>     - Encourage JSR223 Samplers + Groovy  + Caching instead of Beanshell
> >>     - ...
> >>
> >
> > point 2) too
> > The "check script for load test" wizard can displaying warnings for this
> > elements, and give some tips for performance
> >
> >
> >
> >> Maybe we should post this mail on user mailing list to see what users
> >> think
> >> about it.
> >>
> >>
> >
>



-- 
Cordialement.
Philippe Mouawad.

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message