Self-Driving Databases Are Ready to Hit the Road

To advance in the digital economy, enterprises are building and leveraging increasingly complex database systems—too involved for the manual processes data administrators and managers have employed for years to keep things humming. As a result, data managers are turning to automation and autonomous databases. Already, database functions such as backup and recovery are highly automated, and plans are underway to automate such day-to-day functions as monitoring, provisioning, and maintenance.

These are the findings of a new survey of 217 database managers and administrators from the Quest IOUG Database & Technology Community representing a range of industries (financial services, government, technology, and healthcare) and company sizes. The survey, conducted by Unisphere Research, a division of Information Today, Inc., in partnership with Oracle, covered a broad sample of company types and sizes.

With the emergence of data-intensive activities such AI and the Internet of Things, workloads are getting heavier for data managers. Just about all data managers, 95%, have seen increases in data volume over the last 3 years and expect this trend to continue, the survey found. More than nine in 10 respondents are finding it difficult to keep up with this growth. Three out of every four DBAs manage more than 10 databases, with some handling hundreds.

Automation and autonomous databases are a welcome development. Three in four DBAs feel that applications can be deployed faster with increased database management automation.

What aspects of database management are now already highly automated? By far, backup and recovery functions are the areas that have seen the greatest levels of automation, with close to half of respondents, 48%, reporting high levels of automation. Database monitoring is also a highly automated area, cited by 43%. Enterprises intend to step up their automation of database backup and monitoring processes. Database security will also increasingly be automated as database managers continue to brace for the onslaught of threats that accompany the move to digital.

Autonomous databases—self-driving databases that are constructed as cloud databases that use machine learning to eliminate the human labor associated with database tuning, security, backups, updates, and other routine management tasks traditionally performed by DBAs—are ready for prime time. Close to seven in 10 respondents, 68%, say autonomous databases are ready to assume basic database tasks. A majority—more than nine out of 10—indicate that at least some of their autonomous database capabilities will be delivered via cloud services.

Data managers welcome the advancement of automation in these areas, and see greater roles for themselves in higher-level business decision making. Autonomous environments translate to more meaningful job opportunities. Database managers and administrators don’t fear the increased use of automation and machine learning in database operations. The majority of respondents, 60%, expect the arrival of autonomous databases to enhance or elevate the roles of DBAs within their enterprises.

Autonomous databases will deliver solid business benefits. More than 60% of respondents expect their organizations to achieve greater innovation and faster time to market as they automate. There are benefits seen by IT departments as well. A majority of respondents to the survey, 55%, anticipate fewer database errors and bottlenecks. Greater automation will lead to more cost-effective database rollouts, 41% of respondents indicate.

That doesn’t mean humans will be left out of the equation entirely—their roles will expand to oversee these systems as well as to ensure delivery of expected results to their businesses. There is a recognition that autonomous dat bases still need to work in concert with humans. If anything, autonomous data environments will actually be a blend of machine and human expertise. A majority of respondents, 59%, believe there will be increased human oversight as their autonomous databases are implemented.