September 15, 2016

Eliminate the DevOps Blind Spot with Database DevOps

You’re bought into DevOps. You may have purchased Application Release Automation software. Your application development teams are happy. Now, when you read about the benefits of DevOps, you say to yourself, “That’s me!” You should be proud.

But, you’re not done yet.

During your DevOps planning meetings, when you evaluated tools, when you empowered stakeholders to control their destiny through DevOps…did you invite the database team? Have you thought about database DevOps?

We get it. The Database is easy to overlook during DevOps planning. For years, DBAs have protected your company’s most precious asset. To that end, they have created a rigid, manual process to protect the Database. When we were releasing applications once a quarter, that worked great! We wanted the world’s most advanced computer (the human brain) to manually review each SQL script, each proposed change was vetted by a human. Now that DevOps has accelerated the releases and increased our sheer number of applications, you are killing your DBAs. Hence to need to for database DevOps.

Understand this: Database schema changes are performed manually. When your application developers need a database schema change, they create a change request ticket. The ticket may describe a new column to be added to a table and then a PERSON responds to that ticket. The DBA will review the change request, make sure that the request adheres to standards, and then execute the change. Typically, they will manually execute a SQL script. Then, they close the ticket.

In an automated DevOps world, it’s easy to see where the bottleneck is…the Database. This is your DevOps Blind Spot. And just like the blind spot in your vehicle, if you do not manage this…you will get in a wreck. It’s time to bring DevOps to the database.

Your releases are still gated by database changes. No matter how fast you are able to release the application, you will reach a point where you can’t go any faster. A herd can only move as fast as its slowest member.

One major consumer bank that uses Datical to for database DevOps reports they are 300% more efficient than teams that continue to manually make database schema changes. This bank was using a very popular ARA solution to push application changes and found that they were killing their DBAs. The workload had increased to an unmanageable level. They thought they had to make a choice between speed and safety. They got both when they added Datical DB to their DevOps toolchain.

The term they use is “gear ratio” to describe the number of DBAs required for a specific number of changes. In essence, they have moved their DevOps practices into high gear. They are able to move further, faster with less effort. They have eliminated the DevOps blind spot.

To learn more about eliminating the DevOps blindspot with database DevOps, checkout out the DZone DevOps for Database Ref Card.

Share on: