Discussions

News: JTRunner 3.0 Testing Framework has been released

  1. JTRunner 3.0, a JSE & JEE stress-testing framework, has been released. JTR is an open source JSE/JEE testing framework based on inversion of control (IoC). It allows you to write complex stress-test suites for any kind of application ranging from simple JSE ones to complex JEE applications involving EJBs, MDBs and WebServices.

    This new release adds support for testing webservices.
    A new abstract base runner has been added to the JTR framework (AbstractWsRunner).

    This is the abstract class every user-defined runner implementation should extend in order to inherit webservices invocation abilities.

    Finally the jtr.xml configuration file is described by means of an XML schema, so you can validate your own configuration files against the jtr.xsd.

    The webservices invocation component is pluggable: the JTR framework defines a set of interfaces you can implement if you want to change the default webservices invocation enginge which is based on WSIF and DII (Dynamic Invocation Interface). The definition of which concrete webservice invocation engine to use is located at the top of the jtr.xml configuration file in the factories section.

    Read the full project documentation in order to get more information about the number of facilities JTR 3.0 has.

    Threaded Messages (5)

  2. Comparison of JTRunner Vs JMeter?[ Go to top ]

    Anybody got a comparision of JTRunner's features compared to Apache JMeter?

    Paul
  3. Comparison of JTRunner Vs JMeter?[ Go to top ]

    Anybody got a comparision of JTRunner's features compared to Apache JMeter?Paul

    Hi Paul,
    firstly I have to humbly admit that I only superficially know JMeter, and maybe I could be accused of conflict of interest posting my opinion, but I promise I am going to be as much fair as I can :).

    Essentially JTRunner (JTR) gives the developer much more control over what will happen during the tests. That's because you have to code your concrete runners. This little extra-effort is fundamental in my opinion, since it enables you to write even complex test suites, and not only simple request/response tests.

    For example, JTR gives you a lot of flexibility in terms of your runners parameterization. You can have different instances of the same concrete runner class acting in the same epoch of your test with different sets of parameters, and there are even different strategies that guide the assignment of those parameters-set to your runners instances.
    You can get more information reading the JTR official documentation @ http://jtrunner.sourceforge.net/#JTR-IRunner_section:_runner

    The same concept applies to MDBs, EJBs and WebServices testing: with no knowledge of what is actually happening under the hood, you can still write complex interactions with those kind of server components. I humbly think this is the main strength of JTR.

    A look at the JTR documentation will help you to clearly understand why I decided to write my own test-engine, but the main reason is that I was looking for control and flexibility combined with ease of use.

    Furthermore I have to admit that JTR lacks the ability to test http resources at the moment, but this limitation will be removed soon.

    Anyway I firmly think JMeter is a good tool, at least because it has been devised by another italian guy ;).

    Hope someone will give a more independent opinion about the topic :).

    Thanks for your interest,
    Francesco
  4. Grinder ?[ Go to top ]

    May be it would be intersting to compare that to The Grinder.
    Grinder 2 supports extensibility, and v3 is backed by Jython scripting, which definitely have advantages to write complex scenario - though i 'd love to see Groovy used in there as well or any java scripting choice thru some pluggable mechanism
  5. Comparison of JTRunner Vs JMeter?[ Go to top ]

    Anyway I firmly think JMeter is a good tool, at least because it has been devised by another italian guy ;).

    I must've been adopted, because I wasn't aware of this!
  6. Comparison of JTRunner Vs JMeter?[ Go to top ]

    Anyway I firmly think JMeter is a good tool, at least because it has been devised by another italian guy ;).
    I must've been adopted, because I wasn't aware of this!

    Sorry, but I just read it on the JMeter User's Manual Introduction. Follows the excerpt and the link:

    "Stefano Mazzocchi of the Apache Software Foundation was the original developer of JMeter."

    http://jakarta.apache.org/jmeter/usermanual/intro.html

    check out the "History section.

    Francesco