Software Engineering Group

Markov4JMeter – Probabilistic and Intensity-Varying Workload Generation for Session-Based Software Systems


The load and performance test tool Apache JMeter doesn’t support the definition of probabilistic user behavior, i.e. given a currently visited Web site, the next site to be requested is chosen based on configurable probability values.

A Markov Chain is a formal model having a number of states and transitions. Transitions between states are annotated with probabilities. The sum of probabilities of outgoing transitions must be one.

Markov4JMeter is a JMeter component which adds a new Controller, supporting the test plan execution based on probabilistic user behavior(s) itself based on the Markov Chain concept (as well as elements known from state machines, e.g. guards and actions). Moreover, Markov4JMeter enables the variation of the number of active parallel user sessions (threads) during test run – to be specified by user-definable mathematic formulae.


Markov4JMeter is licensed under the terms of the Apache License Version 2.0.

System Requirements

Markov4JMeter requires Apache JMeter 2.2 or higher (versions 2.3+, including the current 2.3.4 release, are supported) and Java 1.5 or higher. Note that JMeter 2.3.1 has a bug in the cookie manager!


We provide a tutorial which consists of installation instructions as well as a step-by-step description on how to develop a test plan with Markov4JMeter. It is included in each release. Additionally, we provide an HTML version of the tutorial.


Download Markov4JMeter (binaries and sources) from our Markov4JMeter project homepage at




Upcoming Events

  • May 24, 2021 09:00 AM to May 28, 2021 04:00 PM Virtual ICSE 2021
  • Aug 23, 2021 09:00 AM to Aug 27, 2021 04:00 PM Athens, Greece ESEC/FSE 2021
  • Sep 27, 2021 09:00 AM to Sep 28, 2021 04:00 PM Luxembourg VISSOFT 2021
  • Oct 04, 2021 09:00 AM to Oct 08, 2021 04:00 PM San Francisco IC2E 2021
  • Nov 15, 2021 09:00 AM to Nov 19, 2021 04:00 PM Melbourne ASE 2021
Research Projects