Test data management (TDM) needs to evolve to keep pace with today's development speed requirements. Find out how to rethink TDM with DevOps
This article was originally published on the Delphix website
here June 25, 2024.
In today’s digitally driven markets, business speed practically equals software development speed. As software accrues greater importance in determining a business’s success in the market, adopting modern software development methodologies such as Agile and DevOps have become crucial for accelerating businesses’ software delivery timelines.
DevOps builds on Agile principles to significantly enhance developers’ work by emphasizing automation and tearing down the silos that often slow down development, testing, integration and releases of software. But many businesses have hit a wall with DevOps. As business data grows and evolves rapidly, and as data privacy and security accrue even greater concern, DevOps teams are increasingly finding themselves facing roadblocks related to their application data. It’s no longer their code or even their infrastructure that holds them back; it’s the lack of fresh, production-quality and compliant data that inhibits their ability to release new and updated software.
Unlike many other DevOps processes, test data management (TDM) processes remain manual and high-touch-oriented, putting them at odds with the automation inherent in DevOps. But adopting DevOps test data management (DevOps TDM) can automate and accelerate TDM processes so that they can keep up with other DevOps processes.
The Need for (Development) Speed
For enterprises today, quick software is now required for both acquiring and keeping customers. As innovation steers more and more aspects of our lives towards the digital plane, companies across industries must digitize operations, interactions, and supply chains to stay relevant.
IDC’s 2022 Worldwide CEO Survey found that a striking 95% of CEOs reported that they were pursuing a digital-first strategy. This digitization trend shows no signs of slowing — additional IDC research found that by 2027, 41% of the average enterprise’s revenue will come from digital products and services.
However, not all software is created equal, and neither are the methods in which it’s created. The ways in which software departments build and deploy software have changed drastically over the past 20 years, but TDM has not kept pace.
The Waterfall Era
Until the turn of the century, most businesses used an aging software development methodology known as Waterfall. The Waterfall method functions by having a fixed software solution linearly pass through a sequential series of defined development phases to ultimately produce a successful software update once every six to 12 months or longer.
While Waterfall allows for measurable milestones, it doesn’t allow for solutions to adapt or change as they progress through the model. This can be problematic for dynamic solutions like software products, which may undergo significant changes during those long Waterfall sequences.
Like most software processes of this era, TDM practices were manual, time-consuming (requiring weeks or months), and infrequently implemented. Updating test data once per year was fairly common. For years, this approach to TDM was acceptable because businesses were using Waterfall to produce one software update every few months— if even that often. So, TDM was an inconspicuous part of software development-- its lack of speed wouldn’t stand out until development processes sped up.
Agile: A New Hope for Software Development
The release of the Agile Manifesto in 2001, which yielded the Agile software development methodology, fundamentally transformed the way enterprise software is developed. The Agile methodology delivers working software in short cycles – typically one to four weeks – that incorporate frequent feedback loops and continuous improvement.
Agile welcomes changes in requirements, prizes simplicity, and encourages cross-team collaboration. These traits put Agile in stark contrast with Waterfall and have helped it win over developers worldwide since its debut. Yet even at that time, TDM largely remained the same—manual, high-touch, and slow.
DevOps Changes the Game Further
By the end of the 2000s, software development began to experience another evolution through the rise of the DevOps methodology and cultural movement. DevOps emerged as a backlash against the longstanding yet inefficient business practice of siloing the software developers who create software (“Dev”) away from the IT operations professionals who deploy, run and monitor software (“Ops”).
The results of fusing Dev teams and Ops teams together are that it:
-
Enhances community spirit
-
Establishes better workflows
-
Allows for quicker and greater incorporation of feedback loops.
-
Facilitates the release of software updates daily.
DevOps completely changed the ways software and IT teams thought about development — yet TDM again failed to evolve to match DevOps’ innovation. But this time, DevOps’ many innovative efficiencies made TDM’s slow, manual processes start to stick out like a sore thumb.
Why Slow TDM Hurts Development Speed
There’s a good reason why TDM took so long to evolve with the innovations that Agile and DevOps brought to development. TDM has traditionally been viewed within software departments as a back-office function like data entry, bookkeeping, or claims processing — certainly important, yet far less visible.
So, while TDM has remained a staple within development, many software departments haven’t evolved their TDM processes to keep up with other more accelerated development processes that DevOps enables. Instead, TDM processes are often manual, time-consuming, and infrequently carried out. This approach used to be acceptable because businesses were using Waterfall to produce one software update every few months, if even that often.
But that’s no longer the case. Nowadays, businesses are increasingly using Agile and DevOps to streamline and automate many other software development processes.
The New Data Used in Test Data Management
At the same time, the data that businesses must manage today faces greater challenges compared to the data of previous generations. Today’s data is far more voluminous than the data used in years past. Businesses also face mounting pressure to secure their data against escalating, increasingly sophisticated cyberattacks. And a growing group of data privacy regulations worldwide mandate compliance for businesses’ sensitive data to protect citizens’ data rights.
Put another way, faster processes + greater data volumes + cyber threats + data privacy compliance requirements = a need for efficient data processes. And most software departments’ TDM processes are anything but efficient.
Slow TDM + Fast Development = Bottlenecks
When slow TDM meets speedy development processes, it leads to bottlenecks of data that hinder software testing ability, lengthen time-to-market and hamper adaptability. Many businesses can relate to the difficulties these bottlenecks can cause. 62% of businesses surveyed in Experian’s Global Data Management Report said that an absence of agility in data processes hindered their responses to changing business needs.
The business world moves far quicker now than it did when TDM originated, but their teams still use the same manual, high-touch processes to handle TDM as they did decades prior. Eliminating these bottlenecks requires businesses to evolve their legacy TDM processes to become more automated and account for modern challenges.
DevOps TDM: The Answer to Faster Development
Adopting DevOps test data management (DevOps TDM) is the key to automation and speed. It enables the automation of data security and compliance procedures, as well as manual TDM processes like data provisioning and refreshing. These automations help these processes match the speed of automated development processes.
DevOps TDM updates TDM processes through several core technologies, with the three most prominent being data virtualization, data masking, and connectivity through application programming interfaces (APIs). These technologies work in tandem to solve various issues facing modern-day data processes:
-
Data virtualization shrinks a company’s data footprint and facilitates quick provisioning of test data, which reduces storage costs, improves sustainability efforts, and improves business velocity.
-
Data masking streamlines compliance and security practices by anonymizing test data.
-
APIs allow one to orchestrate and monitor these automated actions from other software, such as development tools and IT service management tools.
DevOps TDM For Dummies® Delphix Special Edition
Find out why DevOps test data management (TDM) is magic in this eBook. You'll explore how DevOps TDM enables speed, quality, and compliance — without manual provisioning or copying of any resource. And you'll get real-world use cases of test data management in financial services, retail, and human resources.
Get DevOps TDM For Dummies >>
The Benefits of DevOps TDM
DevOps TDM delivers numerous benefits for DevOps teams and beyond. IDC surveyed 10 businesses that implemented DevOps TDM via the Delphix Data Platform and found that Delphix yielded them a five-year ROI of 560%, with a payback period of nine months. They also saw improvements in a variety of business areas, including:
-
76% reduction in business risk
-
72% improvement in total cost reduction
-
40% increase in application development productivity
-
39% reduced total cost of data operations
-
24% increase in IT staff productivity
-
1% growth in average annual revenue
The Delphix Data Platform is a comprehensive DevOps TDM solution that leverages data masking, data virtualization, and APIs to ensure automated delivery of lightweight, compliant data wherever it’s needed. Request a demo today to learn how Delphix can meet your organization’s test data management needs.