With digital initiatives across all enterprises, nowhere is the impact of these changes felt more than in Test Data and Test Environment management.
The ever-increasing challenges around data and environments
The adoption of Agile and DevOps has led to shorter delivery cycles and more frequent releases, for which data and environments need to be provisioned. From a data perspective, most organizations in this space are still maturing. Especially for AI- and machine learning-based applications, it’s clear that these applications have a voracious need for data. For example, an AI based on financial risk modeling depends on having the best data from the most recent day’s trading to balance risk and investment decisions. It is often overwhelming to procure the amount of data needed in the short time between the market close and open. From an execution perspective, automation will be a huge challenge since DevOps tool chain automation still hasn’t become pervasive. Many tools are still relying on clunky, high-investment models such as sub-setting, or solving their data delivery plumbing problems (for issues such as synchronicity, distributed referential integrity) by writing more logic.
In my interactions with enterprise clients and data solution providers, there are some ideas worth exploring:
- Reset and re-test: Explore taking an XTB of a dataset, run a destructive test, and reset rapidly
- Sharing is caring: Testers and developers share data. Find a bug, bookmark the before and after of the entire dataset. There are tools in the market to help you do this.
- Repurpose and repurpose rapidly: After sharing the bookmarks with the developer, the test team can reset their environment, start another test, and still be able to go back to that bookmark.
- If there is an API economy, why can’t there be a data economy? Everything is shareable at scale and at speed. QA/Dev can use their data environment to run tests on multiple releases with full-size data sets, swapping them out as needed. When you’ve tested a bug fix, you can temporarily go to your bookmark and then roll back again to one of the releases.
A wise man said data is the new oil
Well, many argue. However, it only establishes the fact that data is a vital resource for an enterprise. Trends such as the growing utilization of containerized test environments, the use of bots for zero-touch automated testing, the creation of tools for solutions for better test data sampling as well as initiatives such as the Swedish Government’s open data project are all positive developments that will help tackle the challenges facing the industry.
One thing is clear, no matter what challenges or solutions come up in the future, test data and test environments are the two areas that the industry absolutely needs to get right if it wants to move forward on other fronts.
If you have any comments or ideas that have worked well for you, please share your views in the comments below. Particularly, practices around managing data for cloud, performance, regulatory needs.
Download the tenth edition of the World Quality Report 2018 to get deeper insights into the latest trends in the world of testing.