How to End the Wait for Test Data

If you’re a tester or a member of the quality assurance team who works with traditional development methodologies, you know that sound test data management (TDM) becomes a key component of delivering test cycles. Hence, agile data delivery becomes a cornerstone for agile test data management practices. So how exactly does agile delivery fit within the TDM framework?

Below is a simplified diagram of injecting secure agile data delivery into the test data management process, representing an agile TDM ecosystem.

Software development companies use testing to check whether the actual application output results match the expected outcome and to ensure that the software system is free of defects and addresses the business needs. It involves executing a software or system component to evaluate one or more test cases for review, which can either be done manually or through automated tools.

But in all cases, the TDM process must be agile and flexible to meet the demands of excellence in software delivery. So how does one go about introducing agility into test data management?

The answer is simple and one you have heard for over a decade – shift DataOps management to the left of the software development phase. In other words, allow the test data delivery to be incorporated into the front of the build phase, not towards the end. You can then create versions of test data similar to versioning software and merge the two as the new environment is materialized for development.

Here are 4 steps to secure _agile data deliver_y for an agile test data management process.

Virtualize Data

Data virtualization gives users access to the data they need, fast. This technology optimizes the heavy resources a database consumes and for storage, allowing for fast and efficient database provisioning much like server virtualization. It creates highly compressed, virtualized data copies from any data source to produce minimal footprints and help increase storage efficiency and lower costs.

Data virtualization also requires a link to an external source, like the Delphix platform, that is non-disruptive, minimally invasive, continuous and efficient. Ideally, this sync system should be kept current to under 24 hours. This allows a time flow history specific to a point in time should a newer copy be needed for refresh.

Mask Data

By implementing data masking at the origin of the data (before it’s copied and distributed to many locations, insecure devices, and uncontrolled environments), you protect sensitive data where it is most vulnerable. I have often read that prison escapes are said to happen most when a prisoner is transported. The similar can be said of sensitive data. Protect it before it is moved from one location to many others.

Provision Data via Self-Service

Self-service data provisioning lets everyone focus on getting the job done rather than wrestling with lengthy approval processes. Self-service provisioning is simply a shift in defining provisioning policies ahead of time and automating them, and this really defines the Agile component in agile test data management. The big benefit here is not only spending less time waiting for provisioned infrastructure but to also reinvesting that efficiency to bring better products and services to market, faster.

Automate Data Delivery

The term automation itself refers to removing manual work, which in turn saves time to complete a job with accuracy and high velocity. Saving time for completing a job is directly related to saving your tester’s time, so that he or she can get involved in other productive tasks rather than doing a job with mechanical repetition of steps.

Furthermore, automation enables QA teams to access their data via self-service and easily bookmark and reset their data to any previous state without relying on an IT ticketing systems. Bold and innovative decision making happens when engineers have the freedom to experiment with data.

Download this whitepaper to learn about how to turn TDM into a critical business enabler and streamline your software development pipelines.