SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 8 of 8
  1. #1
    Junior Member
    Join Date
    Dec 2004
    Posts
    2
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Functional Testing with LoadTesting Tool?

    Is it possible to use LoadTest tool to do functional testing? If not why what is the reason not to do like that? I believe even in load testing we will see the functionality? If i am wrong can you please give details?

  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: Functional Testing with LoadTesting Tool?

    Sure, you could do some functional verifications using your load test tools by verifying certain data in server responses. But this is not what they are designed for and their features will most likely not facilitate easy functional testing.

    Some performance/load test tools have built in functional verification abilities to make this much easier. JMeter has added a lot in this area recently.
    Corey Goldberg
    Homepage: goldb.org
    Twitter: twitter.com/cgoldberg
    Google+: gplus.to/cgoldberg

  3. #3
    Moderator
    Join Date
    Aug 2001
    Location
    NC
    Posts
    6,018
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: Functional Testing with LoadTesting Tool?

    Most functional tools operate at the application layer of the OSI. Most LoadTesting tools operate at the presentation layer of the OSI (the nature of some supported protocols notwithstanding.) There is a gulf.

    That gulf between the conversation which is represented by the performance testing scripts and what is represented on the screen is in both GUI processing and application processing. If you think of a client in a client server architecture as a piece of cake it will make it easier to discuss.

    The GUI is the topmost layer of the cake, comprised of the icing layer and the vanilla payer underneath. Why the vanilla layer too? Well, items such as GUI rendering can take a lot of CPU time depending upon the app, witness the availability of high end video cards for gaming, video processing/rendering and CAD/CAE work.

    Next up is the application processing layer. Think of this as the chocolate layer. It is in this layer where we have the discussion of classical "thick client" or "thin client" architecture. A "classical" web client is probably the premier example of the thin client. No javascript, no embedded objects, display only. All logic is handled by the middle tier and further back. On the other hand a classical thick client might look like something on the order of a statistical processing package. Lots of local processing between the time data is received and the time it is received on the screen. Thin client = an oreo cookie layer all the way across. Thick client = deep, dark, rum-soaked, fruit included, highly complex application processing layer. Notice we still haven't hit the presentation layer yet.

    The above two layers of the cake, the icing/vanilla payer and the chocolate layer constitute the application layer of the OSI model. The thinner this layer, the higher the likelihood that you could use one tool for both. But even with thin client architectures we can readily observe functionality and rendering differences between Internet Explorer, Mozilla and Firebird. All of these items need to be handled by a tool that operates at the top of the GUI layer, sampling that frosting for consistently high quality, checking the contents of a list box, etc...

    Next up (or down the cake as it were) is the presentation layer, represented by a layer of red velvet cake. At the top of the red velvet cake layer where it connects with the chocolate layer we have the classical application layer protocols such as HTTP, DBLIB, OCLIB, FTP, etc.... As we move to the bottom of the layer, where it connects with the cake plate we are looking at the Sockets interface. The plate, the table, the floor, etc... all represents network infrastructure and is outside of the scope of application testing tools and is the home of network tool vendors like Ixia and Spirent.

    Do you pull data from the registry settings, local .ini files or local data to govern how the interface is represented or displayed? I think you see the inherent challenge of trying to determine whether the interface is functionally sound from just monitoring the client-server conversation.

    There is also the problem of where performance testing fits in the overall QA effort. Consider the following phrase, "If it will not work for one, then it will not work for many." This is the premise under which most performance testing tool operate. As such, most are ill equipped to handle classical front end QA related tasks, such as "does my list box contain the contents of every US state, territory and protectorate?" You would wind up writing so much code to address this type of validation that you could have easily justified the purchase of a tool whose efficiency is a better match to the question you are trying to address.

    Could I build a pyramid with a pocket knife? Given enough time and small enough rocks perhaps. It would be horribly inefficient and probably pretty frustrating for the worker and the pharoh who's waiting for his tomb to be completed. Consider seperate tools. In doing so you will arguably improve your efficiency in testing for functionality and performance related items and hopefully keep your local pharoh happy too.

    Hope this helps,

    'Pulley
    .... a change agent
    James Pulley

    Replace ineffective offshore contracts, LoadRunnerByTheHour. Starting @ $19.95/hr USD.

    Put us to the test, skilled expertise is less expensive than you might imagine.

    Twitter: @LoadRunnerBTH @PerfBytes

  4. #4
    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: Functional Testing with LoadTesting Tool?

    Most functional tools operate at the application layer of the OSI. Most LoadTesting tools operate at the presentation layer of the OSI
    <font size="2" face="Verdana, Arial, Helvetica">Most [web] load testing tools operate by sending HTTP, which is an application layer protocol (in both the OSI Reference Model and the TCP/IP Network Model).

    Most functional tools operate by using the client GUI, which is not represented in any network reference model.
    Corey Goldberg
    Homepage: goldb.org
    Twitter: twitter.com/cgoldberg
    Google+: gplus.to/cgoldberg

  5. #5
    Junior Member
    Join Date
    Jun 2004
    Location
    US
    Posts
    26
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Functional Testing with LoadTesting Tool?

    James's description is making me hungry. I'm ready for some cake and ice cream!
    I would say if the load test tool works for what you are trying to accomplish use it, if not look elsewhere. We use load testing tools for non-load test purposes on a regular basis, mostly for taking network traces of long use cases, which are a pain to execute manually.

    Dale
    www.feiste.com - Extreme web analyzer

  6. #6
    Senior Member
    Join Date
    Jan 2005
    Location
    Aurora, Ont., Canada
    Posts
    1,174
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Functional Testing with LoadTesting Tool?

    Dale, if you are testing the application server, for example, you can do a functional test using a load tester because the operation of the GUI is not a consequence for that purpose. Sometimes it can even be an advantage to bypass the GUI so as to isolate parts of a chained process, as was so delectably explained by James. That is why I am always reluctant to post to open-ended queries like "can it be done" because the answer should invariably be "well, that depends". I think that people need to spend more research than to simply post questions to a forum and hope somebody else has done their homework. Now, if you can describe exactly what you are trying to do and then turn to the forum to ask for comments based on prior experience, then the answers will be much more specific.
    Frits Bos, PMP
    frits_bos@hotmail.com

  7. #7
    Junior Member
    Join Date
    Jun 2004
    Location
    US
    Posts
    26
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Functional Testing with LoadTesting Tool?

    The posted question wasn't mine, and I agree with James's "delectable" analysis. I was simply giving an example of when a load test tool might be used for functional testing.

    Dale
    www.feiste.com - Extreme web analyzer

  8. #8
    Moderator
    Join Date
    Aug 2001
    Location
    NC
    Posts
    6,018
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: Functional Testing with LoadTesting Tool?

    The cake model works exceptionally well when you bring in a multi-layer cake after lunch when you're going through your response time results and analysis.

    People always seem to want to hear about my load test results. Hmmm, is it the results or the free cake. I'd place my money on the free cake. Want to get a laugh, snap a bitmap of the app and have it placed in the frosting of the cake. Gets smiles every time.

    'Pulley
    ... a change agent
    James Pulley

    Replace ineffective offshore contracts, LoadRunnerByTheHour. Starting @ $19.95/hr USD.

    Put us to the test, skilled expertise is less expensive than you might imagine.

    Twitter: @LoadRunnerBTH @PerfBytes

 

 

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 10.00%
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 08:05 PM.

Copyright BetaSoft Inc.