Performance Testing

Performance test for software is done to determine the time taken to get a response from the time it was requested under a tolerable work load. This will help to verify other factors like scalability, reliability, and resource usage.

Performance test helps in finding out whether or not the system meets its requirements in terms of handling more users. Further, it also helps to verify whether it functions correctly under anticipated traffic levels.

Business Challenges

• Ascertaining the application's performance under predicted user loads, prior to deployment
• Identifying performance issues, bottlenecks, their causes
• Deploying a tool that overcomes the limitations pertaining to desired performance benchmark

What We Offer

Kumaran delivers Performance Testing methodology to help solve the above issues. Our methodology includes several steps: identifying the best scenarios that cover maximum functionality of the application and number of users, preparing scripts, scripts execution, and identification of bottlenecks. Our engineering team and database administrators will monitor performance test execution and provide solution for the bottlenecks.

Software Development Lifecycle

The above diagram shows the different stages in development. While the cycle arrives at testing stage, application further moves to three sub stages, Functional Manual, Functional Automation, and Performance.

Functional Manual

Manual testing is the base for any automation test. In this stage the application will be verified with requirements/test cases and it is necessary that the applicacion is stable enough for successful automation tests. This test result will give a clear picture on the status of the application requirements.

Functional Automation

Tools can work much faster and accurate than humans. This reason makes regression testing easier with tools. The scripts that were created in the previous version will be executed to verify functional requirements in the new version. This test gives the status about any existing functionality breaks.

Performance

The above two tests are required to start a performance test. During this test the performance tester will collect application and user statistics to create test plan, scenario, and scripts.

The statistics and benchmarking will vary depending upon the project and environment. The test will determine whether the application can meet the expected load.

Our Approach Towards Performance Testing

The performance test will be done on a certification environment called CERT. This will have a stable application that has passed the functional and regression tests and the environment, which replicates the production environment.

Performance validation (PV) is done before moving any new application or release to the production stage. The various stages involved in Performance test are PV Request, Planning, Scripting, Shakedown, Execution, and Analyzing.

Table of Performance Testing Stages

Performance Testing Framework

KSTS performance testing services include:

Load Testing

Load testing determines the behavior of the application, as a whole or by parts, under a reasonable and expected load. This test will give the response time of each transaction with a breakup of duration for each component that is called within the transaction.

Stress Testing

This test is done to identify components of the software application that tend to fail when an unexpected load is introduced in the system.

Soak Testing

This test is carried to ensure that an application can sustain operating for a period of time. It is different from load testing in that this test continues the iteration until a specific time.

Business Benefits

• Provides confidence in the performance of the applications before deployment
• Mitigates risks associated with the performance and capacity of a new application
• Identifies application performance bottlenecks that impede intended business functionality and their causes, also provides detailed evidence that will speed up the resolution of problems
• Establishes performance baselines for service measurement and provide information prior to launch, which optimizes infrastructure efficacy and supports successful service operation management
• Reduces service availability risk and prevent loss of revenue for customer