The State of Database Deployments in Application Delivery
To account for growing customer demands, relentless competitive pressures, and non-stop market changes, companies are moving faster than ever to release applications. To speed up application delivery, many development organizations rely on DevOps, Agile, and Continuous Integration/Continuous Delivery practices and tools. But faster application development and shorter release cycles also mean more frequent database schema and logic changes. While the application release process has been fast-tracked through modernization and automation, the database release process has been neglected.
In many teams, database administrators (DBAs) still spend hours, sometimes days, manually reviewing database changes, script by script. By employing a slow, manual, error-prone process to update the DB, application delivery is slowed down, no matter how much the rest of the application release process is sped-up. This bottleneck has a significant impact on the business. So, how much of a problem are database deployments for today’s enterprises? Who experiences it the most? And would better automation of database deployments make it easier for companies to release new application innovations faster while reducing risks?
The following key findings are taken from the 2018 Dimensional Research survey, The State of Database Deployments in Application Delivery, sponsored by Datical.
Database deployment create challenges for application teams
One of the implications of faster application deployment and shorter release cycles is that database changes must be pushed out at a faster rate. Not only that, but DBs must be pushed out while maintaining quality and safeguarding company data. When application release stakeholders working at large companies were asked about their database deployment process, 96% say they need to do it faster. But this is challenging. Most (86%) report that this is difficult to do. This includes 40% that characterize their challenges as either “extremely” or “very” difficult.
More than half of all significant application changes require database changes
To quantify how many application changes need corresponding DB changes, we asked stakeholders about the last 10 times they made application changes. An overwhelming majority (71%) report that at least half of all significant application changes also need changes to the database. Also, from a productivity standpoint, most database changes are not one and done. In fact, 91% of stakeholders say they have to re-work database changes many times to get them production-ready.
Database deployments get harder as application release cycles get faster
Today’s enterprises are under pressure to speed up application delivery regardless of industry or company size. The overwhelming majority of application release stakeholders (90%) say they face pressures to release applications more quickly. This pressure is due to the need to respond faster to customer demands and market changes.
When we drilled down on specific steps in the application release process, one of the biggest costs of database deployments are highlighted as a common bottleneck. But what is even more interesting is that the DB release process becomes more of a bottleneck as release cycles tighten from months to weeks. More than half (57%) of application release stakeholders say that DB releases create bottlenecks for application release cycles. They also say that these bottlenecks may last days or even weeks.
Automating database deployments provide tremendous business value
Even though application releases are largely automated and moving at rocket speed, there is hope for database releases. DB release automation can help enterprises shorten the time it takes to deliver application updates to market.
DB release automation and general best practices help database deployment get rid of:
- Security vulnerabilities
- Data loss
When asked about automating the database deployment process, 99% say it would be helpful to their dev organization. Also, 91% of those surveyed say database deployment automation would speed up their application release cycles.
About the survey
These findings are based on an online survey of 303 application development, application release, and DevOps stakeholders responsible for mission-critical applications that connect to enterprise databases. The goal of the survey was to capture hard data on real-life experiences with database release.
To learn more about the States of Database Deployments in Application Delivery and see more findings checkout the survey report.