Performance tests with SoapUI

When wanting to perform performance tests, we evaluate what tools we can use. Among the best known, we can find:

  • Jmeter
  • LoadRunner
  • Neoland
  • LoadUI

However, if SoapUI is used to perform functional tests, we can perform basic performance tests focused on the load.

This content was written by one of our employees, who helps us build our amazing team. Show off your wisdom and join our team!


Before starting with the explanation or showing how to perform the performance tests contemplated in SoapUI, it is necessary to review the concepts of performance tests to know which of these tests are being carried out with SoapUI and which are the ones that we need to be able to choose another tool or at least have identified what tests we are doing with SoapUI.

For what purpose are performance tests performed?

A performance test consists of a load simulation to analyze the performance of the system during the test.

Performance tests are performed in order to identify.

  • Bottlenecks: In other words, if the system collapses, it can be detected if it is due to a lack of CPU, memory consumption, insufficient bandwidth.
  • Identify and locate performance problems: Based on this, adjust the application to perform the tests again to validate that what has been found does not happen again.
  • Delay in response times: For example, if it is established that the application has to respond to mass consumption (for example, 1000 users) concurrently with a time less than 3 sec. It will seek to create a test under these conditions to evaluate the response.

Performance tests are necessary in order to reduce risks for production start-up. It is important to emphasize that performance tests should not be performed at the end of a project because it will be too late there. Ideally, it should be done during the course of the event, including prior to deployment over a pre-productive environment.

What types of performance tests exist?

Load tests: The idea of this type of test is to simulate the reality to which the system will be exposed when it is in production. It is a nominal value that must be established with prior analysis and collection of information from the daily transactions it receives the system. Also, identifying if they are concurrent or not.

Stress tests: This test should be used to identify at what point the system is “broken.” In other words, here, the load must be greater than expected than the load that was defined as nominal.

Resistance tests: This type of test seeks to see how the system responds when a moderate load level is provided for a long period of time.

Performance tests with SoapUI

To set a load test in SoapUI you must choose this option:

What kind of load tests do we have in SoapUI?

1- Simple strategy

This type of test is the nominal one. Also, the so-called baseline is the one that we will take as a reference, that is, in which we present a normal load.

As can be seen in the figure in this test, five threads will be executed with a specified delay (1000 ms), the Random field specifies the variation in the time that may exist. In this case, a delay of 1 sec or 0.5 sec.

2- Variable Load

In this type of test, the number of threads over the duration of the test varies.

For example, if we start with 20 threads, setting the interval to 60 and the variance to 0.8 (this is the indicator as it has to increase/decrease the threads), we will see that the number of threads will increase from 20 to 36 in the first 15 sec. Then, you will return to 20 threads to finally continue with 4 threads.

3- Change of Threads

In this type of test, the subprocesses are increased in the indicated interval to obtain the response statistics of the application as the subprocesses are increased.

4- Burst

This type of test refers to resistance.

During the configured delay (60 sec) nothing is done. Then, the number of threads (5) is executed so that they last the time indicating in the burst duration (10 sec). This way, you can evaluate the recovery of the application.


This content was written by one of our employees, who helps us build our amazing team. Show off your wisdom and join our team!

You may also like

Important Networking Tools in Development

manual testing

Good Manual Testing Tips

Tools for Testing Software

Three Tools for Testing Software During Development