SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 7 of 7
  1. #1
    Member
    Join Date
    Sep 2006
    Posts
    74
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Web 2.0 and AJAX performance testing strategy

    I was pretty late to realise that LoadRunner and Performance Center 9.0 was out and when I finally got to view one of their webinars for 9.0, I realised that the accent was heavily on Web 2.0 and AJAX based applications.
    Logically analyzing the growing number of AJAX based and Web 2.0 applications and the SOA based architectures, and the vast way in which they differ from the current age web based applications, it seems that a markedly different strategy must be adopted to test the performance of such applications.
    I feel I am not posting a specific question here but am throwing up a topic for discussion.
    What would be the important factors that have to be considered when testing an Web 2.0 or AJAX based - application?
    If required to test iGoogle or a similar application, how would I go about doing it?
    I am sure there are already a couple of white papers on 'How to test Web 2.0 applications for performance' out there - but its sad that I havent seen them myself yet. So if anyone else has, please be kind enough to share them with me. If not, it will be great if someone can throw more light on this topic!

  2. #2
    Moderator
    Join Date
    Sep 2001
    Location
    Boston, MA
    Posts
    4,348
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Web 2.0 and AJAX performance testing strategy

    sowmitr,
    yes, AJAX/DHTML does differ quite a bit from Web 1.0 applications where every form submission or user click causes a page refresh.

    With AJAX, you have lots of small asynchronous requests going on in the background, shuttling XML back and forth from your server.

    So obviously when modeling your user scenarios, you must take this into consideration and create scripts that mimic this behavior.

    However, one great thing about SOA and web services is that they are made for machine to machine communication (as opposed to regular web which is made for machine->human consumption). This makes dealing with a web services more natural as you have a proper API to work with.

    I have had a lot of luck by testing services in isolation. In many circumstances you can create small test harnesses to talk to your service and test them individually. this allows you to get throughput and latency data for individual services (before doing a larger integration test).

    anyone else have thoughts?
    Corey Goldberg
    Homepage: goldb.org
    Twitter: twitter.com/cgoldberg
    Google+: gplus.to/cgoldberg

  3. #3
    Moderator
    Join Date
    Aug 2000
    Location
    Vancouver, BC, Canada
    Posts
    1,189
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Web 2.0 and AJAX performance testing strategy

    The way I look at it that Web 2.0 or 1.0 all result in traffic to the servers. Once we understand this traffic we can simulate this. Like Corey says, it is now smaller packets and traffic gets sent in the background. Still it is traffic and really not that much different from what we have been doing all this time.

    The difficulty as far as I am concerned comes with correctly measuring the user experience. Given the asynchronous nature of AJAX, some of the response times are important to the user experience and others less so, because they really do not delay the user.

    This fact pushes us to deeper understand how the application works, what the factors are that drive the response times as experienced by the user and measure those.
    Roland Stens

  4. #4
    Member
    Join Date
    Sep 2006
    Posts
    74
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Web 2.0 and AJAX performance testing strategy

    Yes Roland, this was what I wanted to understand - how will we measure 'user experience' because ultimately that is what will be our ultimate aim. Given the 'asynchronous' nature of applications, we will definitely move away from page loading times to component loading times (e.g. in the case of igoogle or any widget based page). I have not had the opportunity to try out LR 9.0 but does it or any other load testing tool give the breakdown of component wise load times for a page? If it does not, how else can we measure it?

  5. #5
    Advanced Member
    Join Date
    Dec 1999
    Location
    Toronto
    Posts
    794
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Web 2.0 and AJAX performance testing strategy

    [ QUOTE ]
    I have not had the opportunity to try out LR 9.0 but does it or any other load testing tool give the breakdown of component wise load times for a page? If it does not, how else can we measure it?

    [/ QUOTE ]Even in the previous century this feature was already implemented in LoadRunner.
    Yury
    Testing, Performance Testing, Performance Engineering

  6. #6
    Member
    Join Date
    Sep 2006
    Posts
    74
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Web 2.0 and AJAX performance testing strategy

    Yury, thanks for enlightening me - I am very much aware of the web page component breakdown statistics graph. I am not such a newbie to load testing as I seem to be with expressing myself (I have come to realise that of late [img]/images/graemlins/smile.gif[/img])
    What I said was more in the context of SOA based applications which obviously will be pulling services from many third parties (this is what I called components). The kind of breakdown that the current tools give is only from a machine->human consumption as Corey puts it. Is there any capability lying out there that will allow to measure machine->machine communication without having to build separate test harnesses etc? If not, does LR 9.0 support it?

  7. #7
    Advanced Member
    Join Date
    Dec 1999
    Location
    Toronto
    Posts
    794
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Web 2.0 and AJAX performance testing strategy

    [ QUOTE ]
    What I said was more in the context of SOA based applications ... Is there any capability lying out there that will allow to measure machine->machine communication without having to build separate test harnesses etc? If not, does LR 9.0 support it?

    [/ QUOTE ]I completely agree with Corey_G, that SOA based applications are the easiest to load test.
    Typically, building a test harnesses for web services is a rather simple task.

    At the same time you do not need a custom tool, in case you already have LoadRunner.
    Even in the previous century LoadRunner was able to support HTTP based web services. [img]/images/graemlins/smile.gif[/img]
    Yury
    Testing, Performance Testing, Performance Engineering

 

 

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 07:52 PM.

Copyright BetaSoft Inc.