For years, discussions surrounding digital sovereignty have predominantly revolved around infrastructure. However, the focus is now transitioning to a more critical enterprise layer: the database. Geopolitical pressures, particularly in Europe, are compelling hyperscalers like Amazon and Microsoft to make substantial investments to comply with new policies and regulations. This shift in data governance is prompting organizations to reassess their dependence on managed cloud services. Rather than accepting vendor lock-in, an increasing number of enterprises are viewing assets like PostgreSQL as a portable, cloud-neutral foundation, ensuring consistent behavior across on-premises, private, and public cloud environments. This evolution is sparking interest in what some engineers refer to as Sovereign DBaaS: database platforms that offer cloud-level automation while maintaining control away from hyperscalers.
Leading the Charge
At the forefront of this transformation is Gabriele Bartolini, VP and Chief Architect of Kubernetes at EDB. A well-respected figure in the open-source PostgreSQL community, Bartolini’s credibility is bolstered by his co-founding of 2ndQuadrant and the establishment of both the Italian PostgreSQL Users Group (ITPUG) and PostgreSQL Europe. He is also a co-founder and active maintainer of the CloudNativePG operator and the creator of Barman, a pivotal disaster recovery tool within the Postgres ecosystem. His contributions have significantly advanced PostgreSQL’s status in cloud-native environments, including leading the initiative that positioned EDB as the first Kubernetes Certified Service Provider for PostgreSQL.
Bartolini asserts that this shift is not about compromise but rather about redefining the problem to achieve the best of both worlds: convenience and control. “True sovereignty starts with the database. If your PostgreSQL isn’t portable across environments, you don’t really control your stack,” he explains. By ensuring consistency across environments, enterprises can standardize their deployments, enforce policies, and manage complex, resource-intensive workloads with confidence. This architectural choice also enhances negotiating leverage, regulatory compliance, and long-term strategic flexibility.
- A tempting shortcut: Bartolini cautions that while managed cloud services offer speed and simplicity, this convenience often comes at the expense of control. “Convenience is the cloud’s biggest shortcut, but convenience isn’t sovereignty. Real control means you can move your database anywhere and it behaves the same.” This warning is particularly pertinent for organizations reliant on consistent behavior across hybrid environments, where missteps can lead to operational or compliance risks.
- The leverage play: For many leaders, the transition away from managed services is rooted in long-term leverage. Bartolini frames the initial investment as a strategic trade that secures future freedom and negotiating power, noting that this approach is gaining traction to the extent that hyperscalers are beginning to acknowledge it. He cites a recent Microsoft video encouraging customers to run self-managed PostgreSQL with CloudNativePG on their Azure Kubernetes Service as evidence that portability is becoming mainstream. “As an organization, you gain significant leverage with the hyperscaler because they know you can leave easily,” Bartolini explains. “That portability forces them to provide better offerings and better deals to keep your business.”
Empowering Portability
A key enabler of this portability is the Operator Pattern, an architecture that advances database management beyond mere containerization by extending Kubernetes itself. This approach encodes domain-specific expertise into software, effectively teaching Kubernetes how to manage the entire lifecycle of a stateful application like PostgreSQL. This is exemplified in CloudNativePG, which supports modern microservice database architectures and provides declarative APIs for high availability, backup, and self-healing, utilizing native PostgreSQL streaming replication instead of proprietary cloud tools.
- The operational brain: Bartolini emphasizes that while Kubernetes offers portability, simply containerizing a database is insufficient. The database itself must incorporate operational intelligence and effectively manage its lifecycle alongside the cluster. “By embedding the intelligence of a DBA as an operational brain within Kubernetes, you remove the Operational Wall of the hyperscalers, creating a DBaaS that is automated enough for developers but sovereign enough for the enterprise.”
Concerns about whether this control compromises performance are met with Bartolini’s assurance that the opposite is true. He references upcoming benchmarks demonstrating CloudNativePG on bare metal achieving 30,000 TPS with synchronous replication*, while smaller cloud deployments may only reach 1,500 TPS. Such performance metrics are particularly crucial when considering the demands of sovereign AI.
- The AI cost equation: “The move to bare metal signifies a return to a CAPEX model. With AI, predictable costs are essential. The unpredictability of cloud spending was already problematic, and with AI, expenses can spiral out of control very quickly. Fixed costs are crucial,” he notes. By owning the hardware, organizations can better forecast expenses for resource-intensive AI workloads, avoiding the hidden variability of cloud billing and granting teams greater control over both performance and budgets.
- Bare metal performance: Beyond being a technical choice, transitioning from virtual machines and cloud abstractions can serve as a performance multiplier. “If you move to on-premises and run Kubernetes directly on bare metal, the database can perform as fast as a traditional bare-metal deployment. Many assume VMs are necessary, but deploying on bare metal with local storage provides massive throughput and eliminates inefficiencies like multiple block replication.” For enterprises engaged in resource-intensive AI workloads, this strategy ensures both speed and operational predictability, aligning with the CAPEX-driven cost model that renders AI projects financially viable.
However, technology alone is insufficient for success; a cultural shift is also necessary. Bartolini reflects on how his team of DBAs initially doubted Kubernetes’s potential. He advocates for cultivating a “T-shaped profile” in which DBAs enhance their deep expertise with a broader understanding of Kubernetes. This mirrors the historical adoption of PostgreSQL by innovators such as Instagram, Spotify, and Skype in the early 2000s. The operator’s Custom Resource Definition (CRD) can facilitate this transition, acting as a transparent contract between platform engineers and database experts.
- The evolving DBA: “DBAs have two choices: remain in their comfort zone and ignore Kubernetes, or begin learning.” This decision reflects a broader trend in enterprise IT, where infrastructure and development teams increasingly expect database experts to be knowledgeable about cloud-native environments rather than working in isolation.
- Facing the challenge: Bartolini expresses optimism regarding DBAs’ adaptability, sharing that at EDB, “We’ve shown that with about a month of study, a DBA can earn their CKA certification and develop that T-shaped profile, enabling productive discussions with both developers and the infrastructure team.” This approach not only enhances collaboration across teams but also positions DBAs as proactive architects of both application and infrastructure workflows, rather than passive custodians of data.
“The database team cannot drive this change alone; otherwise, it becomes a ‘Transformation with a capital T’ that fails.” Creating what he terms a “sovereign bubble” often necessitates disconnecting other critical layers from provider-specific services, addressing various aspects from compliance to disaster recovery. Bartolini identifies observability as a significant gap impacting enterprise teams, cautioning, “If your logs and metrics are confined to a provider’s proprietary tool, you are not independent. The change must align with the direction the entire infrastructure is heading.” He advocates for a focus on standard formats and technologies as a foundational principle, rather than relying on a CNCF-native observability stack, to avoid being locked into proprietary tools that restrict transparency, hinder collaboration, and limit the organization’s capacity to scale and innovate independently.
Get your free copy of the O’Reilly book: Building a Data and AI Platform with PostgreSQL
