The online community for software testing & quality assurance professionals
 
Active Topics Today's Topics
Sponsors:
Lost Password?

Home
BetaSoft
Jobs
Training
News
Links
Downloads

News Group:
software.testing


Testing
Automation
Performance
Engineering
Miscellaneous
Statistics
Poll
  QA Forums
  Performance & Load Testing
  What Performance testing tool does ?

Post New Topic  Post A Reply
profile | register | preferences | faq | search

UBBFriend: Email This Page to Someone! next newest topic | next oldest topic
Author Topic:   What Performance testing tool does ?
Loga
Member

Posts: 85
Registered: Mar 2001

posted 02-19-2003 09:39 PM     Click Here to See the Profile for Loga   Edit/Delete Message Copy This Message   Reply w/Quote Search for more posts by Loga Visit Loga's Homepage!
I am interested in learning about the way a performance test tool works

I know what the tool does ? But how it does ?

I know that the tool simulates virtual users to stress the system . But how it is simulating VUsers ?

Any points ?

------------------

IP Logged

jonboy
Member

Posts: 7
Registered: Feb 2003

posted 02-20-2003 04:00 AM     Click Here to See the Profile for jonboy   Edit/Delete Message Copy This Message   Reply w/Quote Search for more posts by jonboy
Hi,
I'm sure other forum members will be able to provide you with much more detailed (and accurate) answers than this, but at the most basic level; a load testing tool sits in between the client and the web server and intercepts all outgoing http traffic and incoming html. Vusers are simulated by simultaneously replaying the requests for information that were recorded. This is a fundamentally different approach from functional testing - driven by simulated keystrokes and mouse actions - load testing merely simulates the information that would have sent had the actions been taken.
Well there's a start - who want to take up the baton?!

Cheers,


Jon

------------------

IP Logged

RSBarber
Moderator

Posts: 991
Registered: Jul 2002

posted 02-20-2003 06:41 AM     Click Here to See the Profile for RSBarber   Edit/Delete Message Copy This Message   Reply w/Quote Search for more posts by RSBarber Visit RSBarber's Homepage!
jon, very well put!

1) Read the FAQ on this forum. There is an entire section about your question and it points to threads that should help you.

2) Thing of it this way, reach down and grab your network cable... Imagine putting a wiretap on it that can record, modify and playback what it "hears". That is visually how a performance testing tool works. For a functional tool grab your keyboard and mouse cables and think the same thing.

3) After reading the FAQ and what you have already read here, please come back with any questions you may have. There are plenty of people here who can answer your question in as much detail as you are willing to listen to.

------------------
Scott Barber, Sr. Performance Engineer
sbarber@noblestar.com
http://www.noblestar.com
http://www.perftestplus.com

IP Logged

Steve Jones
Member

Posts: 82
Registered: Nov 2002

posted 02-20-2003 06:46 AM     Click Here to See the Profile for Steve Jones   Edit/Delete Message Copy This Message   Reply w/Quote Search for more posts by Steve Jones
OK, I'll take a shot.

Different tools work in different ways, but fundamentally they all do the same thing. As jonboy said, they sit between the client (IE/Netscape) and the web server and capture the traffic (let's use HTTP/HTML for example) between the to. The tools have functions built into the language that they use to write the scripts that will send and receive the HTTP/HTML traffic.

Now, there is all kinds of logic and programming stuff you can do with these scripts to make them do what you want.

Now, how the virtual users work. The virtual users are, as the name states, virtual. They do not exist, you don't need 100 machines to simulate 100 users, that's the point of the tool. What the tool does is runs the code of the script as a process for each virtual user. The code is sending and receiving the HTTP/HTML and performing any logic that you have added to the script. And the tools will allow you to specify how many virtual users you want to run, how long you want them to run, if you want them to all run at the same time or start up on an interval, and a whole lot of other options.

Some tools also allow you to run the virtual user as a thread within a process which allows for faster execution because the users/processes aren't fighting over processor time. They are all part of one process, so once that process grabs the processor, then all of the virtual users can run. There are advantages of one over the other and a lot depends on what you are testing and the environment you are testing in. Use the search tool here on the forum to find out more.

Let me know if this helped at all or if you have a question or problem with what I've said. Sometimes I ramble and end up with an incoherent mess.

------------------
Steve_Jones@SoftHome.net

IP Logged

RSBarber
Moderator

Posts: 991
Registered: Jul 2002

posted 02-20-2003 06:51 AM     Click Here to See the Profile for RSBarber   Edit/Delete Message Copy This Message   Reply w/Quote Search for more posts by RSBarber Visit RSBarber's Homepage!
Nice Steve - ya know, I think about how long it took for me to really get my head around what was going on with a load generation tool and now it's completely second nature to me. To all you out there reading and scratching your heads - don't worry, the lightbulb does go on, and it really is easy once it does.

Ok, this thread is barely started, and I'm already adding it to the FAQ.

------------------
Scott Barber, Sr. Performance Engineer
sbarber@noblestar.com
http://www.noblestar.com
http://www.perftestplus.com

IP Logged

All times are PT (US)

next newest topic | next oldest topic

Administrative Options: Close Topic | Archive/Move | Delete Topic | Top
Post New Topic  Post A Reply
Hop to:

Contact Us | BetaSoft Inc. | Privacy Statement

Copyright © 1997-2003 BetaSoft Inc.


Ultimate Bulletin Board 5.45c