By Vadim Tkachenko, co-founder and technology fellow, Percona

As 2024 comes to a close and we look to the year ahead, one thing is clear - the database landscape is poised for transformative changes driven by advancements in AI, cost optimization pressures, and the growing need for operational efficiency. From the widespread integration of vector capabilities in relational databases to the critical tipping point for data repatriation, organizations are reassessing how they manage and utilize their data. Additionally, the evolving role of database professionals, with Site Reliability Engineers (SREs) taking a central position, will reshape how enterprise database operations are run. 

Here are my top three predictions for the future of database technology and operations in the year ahead:

Vector Capabilities Will Dominate Database Product Roadmaps 

I don't think anyone would argue with the idea that AI will see significant growth over the coming year. However, what's often overlooked are the secondary effects of this technological explosion, along with the rapidly growing ecosystem of infrastructure, tools, and other enablement systems beginning to crop up around it. In the database space, I believe this phenomenon will manifest most clearly in the widespread implementation of vector capabilities across relational database management systems (DBMSs). Six months to a year from now, when we look back, I think it will become clear that PostgreSQL had been an early bellwether for this trend, with the development and growing popularity of pgvector - an extension that adds support for vector data operations and similarity searches to the popular open source database. While purpose-built, proprietary vector database solutions have been popping up over the past couple of years, this extension gives users all the capabilities and support they need to store and utilize vector data directly within PostgreSQL itself. And I predict we will see other relational databases follow suit in the coming year. Whether it be through an open source extension like pgvector, or some form of native integration or strategic partnership, more and more relational databases will be working overtime to make their solutions more suitable for AI. And in every case, I believe that will begin with the addition of vector capabilities. 

The Push for Repatriation Will at Last Reach Critical Mass

Whether or not we'll see a massive wave of data repatriation take place in 2025 is still hard to say. However, I am confident that the year ahead will almost certainly mark a turning point for the trend. Yes, people have been talking about repatriation off and on and in various contexts for quite some time. However, heading into 2025, I firmly believe that we are facing a real inflection point for repatriation - where the right combination of factors will come together to nudge organizations over the edge and towards bringing their data back in-house (to either on-premises or private cloud environments which they control). Chief among those factors is, of course, cost. In a recent survey of IT leaders, 58% of respondents said their cloud costs are too high, while an additional 14% said they were "way too high". This isn't all too surprising when you take into consideration the fact that cloud spending currently accounts for roughly 30% of contemporary IT budgets, and is set to climb going into 2025 (according to research from IDC). This of course comes at a time when companies across the private sector (and tech in particular) are tightening their purse strings considerably. However, even more important, in my opinion, is the steady drumbeat of advancements and innovations we're seeing in the database space that are geared towards enhanced useability, ease of deployment, and of course, automation. Ultimately, the easier it becomes to deploy and manage databases on your own, the more organizations will have the confidence and capabilities needed to reclaim their data (and a sizable chunk of their budgets). 

SREs Will Soon Rise Through the Ranks in the World of Enterprise Database Operations

For the past half-century, database administrators (DBAs) have handled just about everything associated with enterprise database operations. From design and deployment to maintenance and optimization, DBAs have done it all. However, given the growing impact that AI, ML and automation in general are having on the space, it should come as no surprise that the professional landscape is in for some changes as well. To be sure, the knowledge, skills, and expertise of DBAs will remain indispensable to database operations for quite some time. However, it is their role as the de facto rulers of all things enterprise data whose days I believe are numbered.  With the ability to automatically deploy, test, and tune databases already on offer - including the ability to deploy them directly into production as part of CI processes - database operations will only continue to evolve. And that shift will undoubtedly be moving away from routine management and towards more strategic collaborative, and technical roles. As a result, it will be increasingly important for DBAs to incorporate new skills into their professional repertoires - especially those that fall within the realm of data analytics. For those that do, I believe their careers will not simply survive the shift, but thrive as a result. So, whether it be a DBA by another name, or someone new to the field moving directly into the role, I believe 2025 will be the year of the SRE in the world of database operations. Or, more accurately, the year of the DBRE - the database reliability engineer. 

Looking to the year ahead, these changes will reshape the future of data management, offering new opportunities and challenges for businesses and professionals alike. I am excited for what's to come as it will undoubtedly be a pivotal year in the ongoing transformation of database technology.

##

ABOUT THE AUTHOR

Vadim Tkachenko co-founded Percona in 2006 and serves as its Chief Technology Officer. He leads Percona CTO Labs, which focuses on technology research and performance evaluations of Percona and third-party products, designing hardware, filesystems, storage engines, and databases that surpass the standard performance and functionality scenario benchmarks. Under his supervision, Percona has developed and shipped Percona Server for MySQL, Percona XtraBackup, Percona XtraDB Cluster, Percona Server for MongoDB.