Load testing is essential for assessing the performance of an application. Integrating load testing in continuous integration and continuous delivery (CICD) can help in identifying critical performance issues that arises as part of code changes. However orchestrating load testing in CICD brings in additional challenges in terms of setting up test data, load generator and controller machines.
JMeter is one of the most commonly used performance testing tools and it uses Master slave model where Master machine acts as a controller and multiple slave machines act as load generators. Having dedicated AWS EC2 instance/VM as controller or load generator machines can increase the testing cost. In this session we are going to see how to reduce the load testing cost by creating EC2 instances on demand using JMeter, Taurus framework, Jenkins, Ansible and Kubernetes. Once the load test is completed, all the result will be moved to repo/artefacts and the EC2 instances will be shut down.
Aravind Sridharamoorthy is a senior performance test engineer in MLC Life Insurance with 15 years of experience in performance testing. Aravind’s has implemented performance testing CICD framework for multiple projects in Telstra and NBN prior to working in MLC. Aravind has experience in functional test automation using Cyress and he has also implemented contract testing in NBN. Aravind is a certified “Yoga teacher” and conducts community yoga classes in Melbourne and his hobbies include swimming and playing chess.