Performance testing is a technical investigation that is carried out to determine or validate the speed, scalability or stability characteristics of a particular product. Products normally undergo performance testing through all stages of their development, rollout and on-going maintenance. This is to ensure that there are no overlooked errors that could damage a company’s reputation or lead to unsatisfied customers or clients. Some aspects of performance testing could involve diagnostics, validation of application performance, proactive performance monitoring, application capacity planning and modelling.
All of this is carried out to try and ensure improved productivity, increased quality and increased speed to market. In software engineering, performance testing is crucial in order to test the software before it is deployed live. It is tested to see how well it performs in terms of responsiveness and stability under a variety of scenarios that mimic the actual application of the software. This performance testing serves to serve to investigate, measure, validate and verify the attributes of the software and how it interacts with other elements of the system. Performance testing can demonstrate that the system meets performance criteria, compare two systems to find which performs better and it can see which parts of the system could cause the system to perform badly.
There are a number of different types of performance tests that can be carried out on a product or on a system. Each type of testing looks at various aspects of the product and how it will perform under a range of conditions.
1. Load testing
This might be the simplest form of performance testing. This type of test is usually carried out to understand the behaviour of the system under a specific expected load. This load could be the expected concurrent number of users on an application doing the same thing within a set duration. It will give the response times of all the important business transactions and monitor the database and server to see if any problems arise.
2. Stress testing
This type of testing is normally carried out to try and understand the upper limits of capacity within the system. Stress testing checks the system’s robustness in terms of extreme load and helps determine if the system will adequately perform the current load goes above the expected maximum.
3. Soak testing
This type of performance testing is also called endurance testing. It is carried out to see if the system can sustain the continuous expected load. Soak testing monitors memory utilization in order to detect potential leaks and to see if there is any performance degradation. Checks are performed at the beginning and end of sustained activity to determine if there is a variation in performance levels. The goal of a soak test is to discover how the system behaves under sustained use.
4. Spike testing
This performance testing is carried out to see whether performance will suffer or fail, and how it handles surges or dramatic changes in load. Spike testing is conducted by suddenly increasing or decreasing the load generated by a large number of users, and then observing the behaviour of the system.
5. Configuration testing
This sort of performance testing shifts the focus from a load perspective to see how the system responds to configuration changes. This might include experimenting with different methods of load balancing to see how it effects the system’s performance and behaviour.
6. Isolation testing
Isolation testing involves repeating a test execution that resulted in a system problem. This just doesn’t happen in performance testing but this type of test can sometimes isolate and confirm the fault domain.