Difference in response time when done through LR and manually
Mithun
James Pulley
You could have a very large, thick, complex AJAX GUI that is slow on the host being measured and adds 8++ seconds to the overall page render time.
You could have third party add ins to your page which reference third party sites which are not included in your direct to server 2.7 seconds. Have you seen how much time Google Analytics seems to add to a full browser render at times!
You could have extraordinarily large complex state management variables that take a while for your browser and app to properly process, but take a much smaller amount of time to actually transfer to the browser. Be careful how much you place in state.
You could have 25 different toolbars loaded in your browser which is slowing it down. OK, 25 is a little much, with an 8 second delay you might have 2-4 third party toolbars. Add 25 of them and you would be crawling for minutes waiting for a response.
Think in terms of layers. You are sampling at the top of the application layer of the OSI model with your manual observations. There is a delta between the top of the GUI and (presumably) the HTTP layer that you are sampling at with your script. Think of all of the things which are going on in your browser between receipt and full render to the user and how much time that may add to your overall end user response time.
James Pulley, http://www.loadrunnerbythehour.com/PricingMatrix
Do you know the nine layers of the OSI model and how knowing the layers benefits you in performance testing/engineering work?
What, you thought there were only seven. New, at the very top, “Politics, without which nothing happens.” New layer zero, “Power, without which nothing moves.” ;)
--
You received this message because you are subscribed to the Google "LoadRunner" group.
To post to this group, send email to LR-Loa...@googlegroups.com
To unsubscribe from this group, send email to
LR-LoadRunne...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/LR-LoadRunner?hl=en
WebTestGadfly
possible that
your non-LR response time into an IE browser is accurate, and the LR
response time
is inaccurate? I mean, could the LR engine take some shortcuts on
this web app that
it is not telling you about?
-WTG
On Jan 19, 9:06 am, "James Pulley" <loadrunner-li...@jamespulley.com>
wrote:
> host being measured and adds 8++ seconds to the overall page render time.
>
> You could have third party add ins to your page which reference third party
> sites which are not included in your direct to server 2.7 seconds. Have
> you seen how much time Google Analytics seems to add to a full browser
> render at times!
>
> You could have extraordinarily large complex state management variables that
> take a while for your browser and app to properly process, but take a much
> smaller amount of time to actually transfer to the browser. Be careful how
> much you place in state.
>
> You could have 25 different toolbars loaded in your browser which is slowing
> it down. OK, 25 is a little much, with an 8 second delay you might have
> 2-4 third party toolbars. Add 25 of them and you would be crawling for
> minutes waiting for a response.
>
> Think in terms of layers. You are sampling at the top of the application
> layer of the OSI model with your manual observations. There is a delta
> between the top of the GUI and (presumably) the HTTP layer that you are
> sampling at with your script. Think of all of the things which are going
> on in your browser between receipt and full render to the user and how much
> time that may add to your overall end user response time.
>
James Pulley
the results of the test and (2) pleadings of users who have not had training
and become frustrated with LoadRunner and cannot explain what is
happening...Blame the tool. I think someone even wrote a poem about this
phenomenon with LoadRunner a while back, I'll see if I can dig it up.
Which is more likely, the sampling method of LoadRunner has been in error
for 15++ years or that the sampling method/items measured by LoadRunner in
this case is distinct from the items being sampled at the top of the browser
layer? The former would have been a pretty easy target for all of
Mercury/HP's competitors over the years (unless they had the same issues)
versus the latter which is very common among users new to the use of the
tool and have failed to cultivate an architectural perspective on their
client and server side architectures.
James Pulley, http://www.loadrunnerbythehour.com/PricingMatrix
Venkat P
Q: What could be the reason for such high variation in timings ? Am I missing something ?
(If your developer/manager doesnt get it - then go for the One page answer - James' email)
--
WebTestGadfly
for one thing. Could it be the other way around?
The original complaint was that the non-LR
AJAX test playback took 11.2 seconds, but the LR-based
playback took 2.7 seconds.
According to the one-line answer, then, is this
difference caused by the LR-based playback NOT
rendering the page as fully as the non-LR version?
I mean, what is it that the LR-based playback is NOT doing
that could make it 4+ times faster?
-WTG
On Jan 25, 8:11 am, Venkat P <vjpj...@gmail.com> wrote:
> (Start with the one line answer here )
>
> Q: What could be the reason for such high variation in timings ? Am I
> missing something ?
> A: Page Rendering Time in the Browser .
>
> (If your developer/manager doesnt get it - then go for the One page answer -
> James' email)
>
>
>
> On Wed, Jan 19, 2011 at 11:47 AM, Mithun <mithun7...@gmail.com> wrote:
> > We have a .NET AJAX ASPX application.
>
> > I am observing large variation in response times when the application is
> > run through Loadrunner and when checked manually.
> > For Eg. I have a aspx browse page which get some records in a .net data
> > grid control from a underlying SQL server database.
> > Loadrunner gives a response time of around 2.7 seconds for a single user.
> > Whereas when tried manually the data in the page came in around 11.2
> > seconds.
>
> > The LR script does not contain any thinktime enabled.
> > There is only one web_submit_data call and 3 correllated parameters in the
> > transaction.
>
> > What could be the reason for such high variation in timings ? Am I missing
> > something ?
>
> > IE version - 7.0
>
> > Please let me know if any other details are requuired.
>
> > --
> > You received this message because you are subscribed to the Google
> > "LoadRunner" group.
> > To post to this group, send email to LR-Loa...@googlegroups.com
> > To unsubscribe from this group, send email to
>
> - Show quoted text -
Floris Kraak
> The one-line answer would be a winner, but
> for one thing. Could it be the other way around?
>
No.
> The original complaint was that the non-LR
> AJAX test playback took 11.2 seconds, but the LR-based
> playback took 2.7 seconds.
>
> According to the one-line answer, then, is this
> difference caused by the LR-based playback NOT
> rendering the page as fully as the non-LR version?
>
LR doesn't render. At all.
Why do you think it should? LR is supposed to measure the effect of
putting load on a server. It was designed to do this by simulating the
network traffic. Page rendering time has been specifically excluded
for a long time now. Only protocols that specifically implement a
thick GUI client will measure GUI rendering time.
> I mean, what is it that the LR-based playback is NOT doing
> that could make it 4+ times faster?
>
Rendering, of course.
Regards,
Floris
---
'Many people asked me if I was afraid to fly and implied that I should
have stayed home, close to family and friends. I replied that if I had
stayed home, the terrorists would have won.
Unfortunately, my government does not agree with my definition of
winning. They think that living in fear and trying desperately to keep
us all 100% safe while flying is the most effective way to fight
terrorism. It reminds me of a boss that told me he liked it when
people lived in fear of being fired, they worked harder. I told him
being fired held no fear for me. When you live in fear, you do
irrational things - like sending millions of people's shoes through an
xray scanner every day.'
-- Stormy Peters
anita koli
--
You received this message because you are subscribed to the Google "LoadRunner" group.
To post to this group, send email to LR-Loa...@googlegroups.com
To unsubscribe from this group, send email to
LR-LoadRunne...@googlegroups.com
mn
--
You received this message because you are subscribed to the Google "LoadRunner" group.
To post to this group, send email to LR-Loa...@googlegroups.com
To unsubscribe from this group, send email to
LR-LoadRunne...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/LR-LoadRunner?hl=en