SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 4 of 4
  1. #1
    Junior Member
    Join Date
    Sep 2000
    Location
    India
    Posts
    12
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Webload results against real life results

    Has anyone been able to verify the results that Webload provides with the actual application behavior on the Web?
    For. e.g do the round times shown by Webload for a particular load and connection speed reflect under similar conditions for an online application?
    i.e. if Webload gives a roundtime of 10 seconds for a partcular transaction under a particular load and connection speed, will i get the same as response time in real time?

    Thanks,
    Vijaya

  2. #2
    Senior Member
    Join Date
    Dec 1999
    Location
    Chicago,Illinois,USA
    Posts
    2,537
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: Webload results against real life results

    I cannot speak directly to WebLoad on this issue but, in general with these types of tools, you can certainly just eyeball the transaction. For example, if it says that the time from clicking a given link to the "completed loading" of the next page takes ten seconds, you can simply stopwatch the exact procedure you are doing in WebLoad. If you see it takes only two seconds, there might be some issues.

    Having said that, you have to be careful because sometimes this can be misleading. For example, let us say you have a Java applet on the next screen that appears after the link. I have seen some tools report round-trip times of, say, seven seconds. And yet, in reality, the Java applet is still just a gray box because now the client-side is processing so the actual user experience of round-trip time is something like fifteen seconds.

    Since a lot of people balk at the idea of eyeball verification you could also check server logs for certain transactions and look, at a much more granular level, at what is happening.

    I am sorry this did not answer your direct question about WebLoad but I think it applies to these tools in general.

  3. #3
    Senior Member
    Join Date
    Jun 2000
    Location
    MA, USA
    Posts
    127
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Webload results against real life results

    The other thing to take into account when "eyeballing" is to make sure that you are doing an apples to apples comparison. I cannot speak for the other tools, but in WebLoad each iteration for each user is treated as a new user: no connections persist between rounds, the user's caches are cleared etc. To compare what WebLoad is doing in a browser environment you need to do the following in your browser:

    - clear memory cache
    - clear disk cache
    - reload the page

    You should then find that the download times you see by eye and as reported by WebLoad are comparable (allowing for other issues affecting performance such as network availability, server side load etc).

    For connection speed emulation, in our internal testing WebLoad is a good approximation to the line speed selected. This is because WebLoad only sends or receives data in lots of small packets when emulating a slow line, as well as keeping the socket held open for the appropriate timeframe to load the server's connection handling properly. On the other hand, particularly at speeds such as 28.8k and 56k, remember that WebLoad is emulating a line at that speed -- it is not emulating a modem operating at that speed. Modems have additional overhead which reduce their effective capacity below the connected line speed. To emulate modem connections you should measure an actual modem to get an idea of the effective speed it achieves, and then apply that value to WebLoad's connection speed setting.

    Jeff is right, however, to also take care about client side interactions, including time to instantiate system elements outside of the page (e.g. the browser will need to create a JVM the first time it hits an applet; this can take several seconds). Ultimately, one of the questions you have to ask yourself is whether the performance characteristics you are trying to measure include elements that you cannot control, including client side issues such as the user machine's performance, the time taken to render complex images etc.

    Hope this helps,

    Phil Hollows
    VP Technology
    RadView Software

    VP Product Marketing
    http://www.open.com
    Real time threat correlation and security management

  4. #4
    Junior Member
    Join Date
    Sep 2000
    Location
    India
    Posts
    12
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Webload results against real life results

    Thanks a lot for the replies.I am planning to compare the response times with webload and the actual ones measured with real modems.I am setting no caches also.

    It would be a great help however,if there are particular values available, obtained out of experience that say that an increment of app. 'x' over the webload result will make my result realistic for particular modem speed(with conditions of network, client side processing to render the html and image applying, load size etc.).
    If any of you have come to such conclusions, it would be great if you could share them.


 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Search Engine Optimisation provided by DragonByte SEO v2.0.36 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 10.71%
vBulletin Optimisation provided by vB Optimise v2.6.4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging v3.2.8 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
vBNominate (Lite) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Feedback Buttons provided by Advanced Post Thanks / Like (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Username Changing provided by Username Change (Free) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
BetaSoft Inc.
Digital Point modules: Sphinx-based search
All times are GMT -8. The time now is 10:49 PM.

Copyright BetaSoft Inc.