Confessions of the DBA: “DevOps Test Data Requests Are Always at the Bottom of My To-Do List”
When we speak to DBAs, we hear time and time again the requests for data are increasing, and it's becoming near impossible to fit them in amongst everything else. It’s becoming a balancing act that is putting serious pressure on their daily functions, often with some unintended consequences.
Database administrators have one job - protect the data. They manage the performance, integrity and security of databases, but are increasingly expected to help refresh growing, non-production environments more quickly - without hiccups or disclosing sensitive information.
They also need to make sure that data is anonymized and pseudonymized when it’s being copied over from production environments to non-production environments for development, testing and reporting, all the while staying compliant with regulations.
Not to mention, DBAs are accountable for restoring database backups into test environments, which in many instances are prone to error - causing greater frustration and difficulty.
In today’s on-demand economy, testing applications have become hugely important in the development process to deliver software at the pace needed to keep up with changing consumer needs.
Between all these tasks, ranging from testing and validating backups, fixing performance issues to pushing changes, DBAs are asked to support ad hoc requests from the DevOps teams to support growing data demands.
At a large enterprise, the scale of managing data can be overwhelming and can cause tremendous data friction in a company’s digital operations. Because data is more important than ever to drive production, more people across the organization want access to it.
Hence, data friction emerges when the demands of DBAs, developers, data analysts and business decision makers, are not met by data management and IT professionals. DBAs, not surprisingly, are caught in the middle - stretched in both directions.
“When data friction becomes the blocker to innovation, customers leave, competitors win and businesses spend more time reacting, instead of leading,” as Delphix CTO Eric Schrock, writes.
This is why an agile approach to data management is crucial. Historically, application teams have manufactured data for developing and testing in a siloed, unstructured fashion.
Let’s dive into a real world example. The DBA team at GAIN Capital, a global leader in online trading, took a different approach to tackling data friction. They wanted to bring agility to its IT operations teams that translates into faster app development, ultimately bringing greater value to the business.
Application teams worked on nearly eight projects simultaneously, where delivering data environments for a single project took up to four hours. Multiplied over eight projects with regular environment refreshes, it was clear that data had become a barrier to the development process.
GAIN found itself having to sacrifice or delay new, innovative projects in order to prevent delivery rates from slipping.
Once the team decided to shift their approach using new data virtualization technology, the time taken to create database copies was reduced from four hours to just three minutes; more than 30 developers were able to access data via a self-service model, without a DBA or storage administrator; project times dramatically decreased; faster data delivery times increased its environment utilization, allowing the company to remote QA environments and lower infrastructure costs.
GAIN estimates that development cycle times decreased by 75 percent and IT’s output to the business increased by 20 percent.
What’s the takeaway? Faster and more thorough test data management means faster delivery of apps and faster achievement of business goals, and more importantly, no late nights for DBAs.
Tune into our “Confessions of a DBA” webinar to hear more about what modern DBAs are doing differently to meet data demands. You can also learn more about how the Delphix Dynamic Data Platform can make data fast and secure for access across your organization.