Performance Engineering

Performance testing gathers all the tests that verify an application’s speed, robustness, reliability, and correct sizing. It examines several indicators such as a browser, page and network response times, server query processing time, number of acceptable concurrent users architected, CPU memory consumption, and number/type of errors which may be encountered when using an application.

Why should you test the performance of your system? In short, to ensure that it will meet the service levels expected in production, as well as deliver a positive user experience. Application performance is a key determinant of adoption, success, and productivity.

As it can be cost prohibitive to have to solve a production performance problem, continuous performance testing strategy optimization is the key to the success of an effective digital strategy.

What are the different types of performance testing?

There are different types of performance tests; the most important include Load, Unit, Stress, Soak, and Spike testing.

  • Load Testing

Used to simulate the number of virtual users using an application. To reproduce realistic usage and load conditions, depending on response times, this test can be used to identify potential bottlenecks. It also helps the user fine-tune the application architecture’s sizing

  • Unit Testing

Simulates transactional activity of functional test campaign. The objective is to isolate the transactions that could disrupt the system.

  • Stress Testing

Executed to assess how systems behave in the face of maximum peak activity. To this end, the number of users increases significantly/steadily during the test period.

  • Soak Testing

Stress tests and Soak tests are intended to increase the number of users to monitor system behavior but over a longer period. The objective is to observe whether a dense and continuous activity, over time, displays performance degradation and excess system resource demand.

  • Spike Testing

Verify how systems perform when during the scenario where activity is greater than normal. Unlike Stress tests, Spike tests take into account the user count and complexity of the actions being carried out (hence, the increase in the number of business processes generated)

What do performance tests measure?

Performance test can be used to analyze various success factors such as response times or the error potential. With these performance results in-hand, you can confidently identify bottlenecks, bugs, and errors; decide how to optimize the application to eliminate the issue(s).

When should you perform performance tests?

Be it web or mobile, the lifecycle of an application is comprised of two main phases: development and deployment. In each instance, the operations teams expose the application to end-users on the production architecture.

Development performance testing focuses on components (web services, microservices, APIs). The earlier an application’s components are tested, the faster an anomaly will be detected, and the lower cost of resolution.

As the application is assembled, performance tests are increasingly extensive. In some cases, performance tests can be carried out during the production, for example, when it is difficult/expensive to replicate a production environment in the development laboratory.

What are the characteristics of an effective performance test?

Realism and depth of analysis are the two characteristics of “good” performance tests. It is not only a question of voluminous transaction count simulation, but the enablement of a realistically reproduced user experience as if the application were in production.

Performance testing can generate a very large amount of data. The right performance test is the one that promotes analysis that quickly/accurately identifies all performance problems and causes.

With the emergence of Agile development methodologies and DevOps process practices, a performance test must remain reliable while respecting the accelerated pace of the cycles: development, testing, and release. To keep up with this pace, organizations seek to automate their performance testing and choose NeoLoad – the fastest and most automated performance testing tool for test design, sorting, and analysis.

Why automate performance testing? For more agility!

Digital transformation is driving organizations to accelerate the pace associated with the design of new services, applications or features in the name of gaining/maintaining their competitive advantages. Agile development methodologies provide a solution in this context.

Despite ongoing industry adoption of Agile and DevOps, performance testing is a highly-manual process. The goal of every performance tester is not to let it bottleneck their Agile development. The solution – integrating as much automation into the performance testing process as possible.

  • Launching tests automatically within a Continuous Integration context, AND
  • Automating design/maintenance tasks wherever possible

Having complete automation of your performance testing is possible when testing components. However, the human intervention of performance engineers is still necessary when performing complex tests on assembled applications.

The future of performance testing lies in the automation of tests across all stages of the application lifecycle.

How to automate performance testing with NeoLoad?

NeoLoad is the performance testing platform that makes it possible to automate test execution, test task design, update, and analysis.


During performance test design, NeoLoad automates correlation and randomization tasks, allowing you to create tests 10x faster as compared with other tools. It also enables you to import existing functional Selenium test scripts for use during performance tests.


In consideration of Continuous Integration, NeoLoad interfaces with the main CI servers: Jenkins, Bamboo, and TeamCity via tool API. At this point, all custom integrations with the different tools of the continuous deployment chain are possible.

Performance test

One of the main challenges for performance engineers is updating test scenarios when the application being tested, changes. This is particularly true when testing assembled applications. NeoLoad offers an almost complete update automation functionality for scenarios created within the tool. Automation is complete for scenarios imported from Selenium.

Performance testing analysis

The most difficult phase to automate, especially for complete tests performed on an assembled application, as performance problems are due to multiple factors. Identifying their cause often requires human intervention. NeoLoad offers a wide range of performance testing analysis capabilities enabling you to accurately identify performance problem root cause and to quickly do so by isolating critical data.

It is now possible to automate a large part of the analysis in an Agile environment. NeoLoad allows you to define expected service levels (SLAs) used to assign success/failure status to a test. This allocation can be fully automated, allowing for the entire performance test cycle to be automated with a CI server. For example, an Agile team can schedule automatic, non-regression performance tests that will run during night builds. The comparison between the performance test results at the expected service level and the previous build’s test results will be carried out automatically, allowing the automatic integration process to continue if the test is successful.

When to do PE?

NTS’s Performance Engineering Process & Accelerators