All of my articles are free to read. If you don’t have a Medium subscription, you can still read by clicking this link.
Introduction:
As test automation engineers, we often face the challenge of managing test data across various states, such as Created, Pending, Authorized, Paid, Refunded, and Cancelled. This becomes especially complex when our projects scale, and we need to accommodate different regions and environments. In this blog post, we’ll explore a cutting-edge solution that leverages Redis to optimize test data management, reduce test execution time, and improve overall test automation quality. Get ready to revolutionize your test automation projects!
The Challenge:
In the early stages of our test automation project, we relied on UI automated scripts and static data from the Faker library to generate dynamic test data. While this approach worked initially, it quickly became unmanageable as the number of tests grew, causing issues with flakiness, application problems, and uncontrollable execution times. Our customer’s request to set up test data for various regions and environments only exacerbated the situation.
The Solution:
To address these challenges, we decided to separate test data creation into a standalone project, allowing it to run either before or in parallel with our main testing suite. This way, we could maintain an inventory of test data, organized by data state, region, and environment, and persistently store it for future use. After exploring numerous options, we chose Redis as our test data storage solution. Let’s dive into the step-by-step process that transformed our test automation project:
Redis Server Configuration: We set up a Redis server on a virtual machine (VM), allocating predefined memory based on our test data size and opening the required firewall port.
Jedis-Based Data Management: To manage our data store, we built a solution using Jedis, a popular Redis Java client. This allowed us to organize by data state, region, and environment.
Exception Handling and Concurrency: We implemented exception handling, concurrency modifications, and considered various edge cases to ensure the robustness of our solution.
Application Automation Integration: We integrated our application automation code to push (on-demand) and pop data from the Redis data store for each category.
Test Code Design: Our test code was designed to check data availability before each suite, initiating data creation if the minimum required data was not available.
The Result:
By implementing this Redis-based approach, we experienced a significant improvement in our test automation quality. Our new test data management system allowed for more efficient test execution, reduced flakiness, and facilitated seamless management of test data across different regions and environments.
Conclusion:
If you’re a test automation engineer looking to level up your test data management game, it’s time to experiment with Redis. This powerful in-memory data store can revolutionize your test automation projects, enabling you to maintain test data inventory, optimize test execution time, and ultimately deliver higher quality results. Don’t miss out on the benefits of this game-changing solution!
