Data sovereignty is reshaping AI infrastructure

The cloud-first mandate is hitting a hard, physical reality: the uncompromising need for data sovereignty. As generative AI and high-performance computing (HPC) scale across government labs, research universities, and regulated enterprises, organizations are drawing a line in the sand. According to recent Intersect360 Research, the infrastructure debate has officially been settled. Many organizations own their infrastructure, site it locally, and staff it themselves.

The numbers tell an undeniable story of a market moving aggressively on-premises to guarantee data control. Intersect360 Research found that 68.2% of respondents use no public cloud storage at all for active HPC-AI workloads, while 56.3% report zero cloud computing usage for these systems. Crucially, 83.5% agree that data and compute must remain physically proximate, with zero respondents disagreeing. This is not a temporary cost calculation or a transient trend; it is a structural necessity driven by strict regulatory exposure, national security mandates, and the absolute requirements of data residency.

The AI performance penalty

While sovereignty requirements are successfully being met on-premises, a massive new technical challenge has emerged: performance. AI workloads are driving storage architectures toward intense, many-small-file access patterns. Legacy parallel filesystems were originally built for large, sequential data streaming, and they struggle immensely when this access pattern inverts.

The research exposes a widening gap between what sovereign organizations need and what they are actually experiencing:

• The small-file shortfall: Performance on many small files represents the largest satisfaction gap in the entire study, scoring a massive 0.98 deficit between importance (4.18) and user satisfaction (3.20).
• The commodity trait ends: A resounding 92.9% of respondents say that storage solutions are strategically important, board-level acquisitions, rather than simple commodity purchases.
• Absorbing the hit: Buyers are currently choosing to absorb this severe performance penalty rather than compromise on data residency or loosen their sovereignty requirements.

Closing the gap: Sovereign-grade and AI-ready

Organizations should not be forced to trade analytical speed for data ownership. The market is actively demanding on-premises architectures that can deliver world-class AI data throughput and ultra-low latency entirely on the customer’s floor.

This is where the alliance between industry leaders changes the calculus for modern IT leaders. HPE Cray Supercomputing systems, powered by AMD EPYC™ CPUs and AMD Instinct™ GPUs, are proving that sovereign-grade infrastructure and AI-ready performance are no longer competing specifications. This powerful hardware profile is already driving some of the world’s most significant, institutionally owned deployments. These include the Frontier exascale system at Oak Ridge National Laboratory, Setonix at Pawsey Supercomputing Research Centre, and LUMI at CSC Finland.

By pairing HPE’s deep operational lineage in highly classified, zero-trust environments with AMD’s cutting-edge silicon architecture, enterprises can deploy local clusters designed specifically for the unique metadata and small-file demands of modern AI models.

The new blueprint for data control

Data sovereignty requires satisfying three pillars simultaneously: total ownership of the physical equipment, complete control over data access, and local accountability for security. As regional regulations stiffen across Europe, Asia-Pacific, and the Middle East, the transition away from public cloud dependence for core AI assets will only accelerate. CIOs who want to maximize the analytical value of their proprietary data without exposing it to third-party risk must look to partnerships that offer deep regional expertise, flexible financing, and validated hardware architectural depth.

The sovereign AI checklist for CIOs

• Audit remote access: Verify whether remote hardware vendors maintain administrative access paths that bypass your data sovereignty boundaries.
• Map data-compute proximity: Identify latency bottlenecks caused by separating active AI training models from underlying data repositories.
• Benchmark small-file workloads: Test how your current on-premises storage handles AI metadata operations and minor-file access patterns.
• Assess regional compliance: Review emerging data-residency laws in your operational regions that mandate local personnel and local physical siting.

3 quick tips for CIOs

1. Don’t compromise residency for speed: The performance gap in small-file storage is a hardware architecture problem, not an indictment of on-premises infrastructure. Look to modern parallel processing systems rather than migrating back to the public cloud.
2. Vet your vendors’ operational lineage: Look for infrastructure partners with a proven history of operating in highly regulated or classified environments. True security capability is demonstrated through an engineering record, not a product brochure.
3. Plan for total handoff, not just installation: AI systems require continuous optimization. Contract with vendors that provide comprehensive, step-by-step guidance and long-term local support networks rather than transactional sales.

To learn more about how to design an on-premises infrastructure that balances compliance with performance, read the Intersect360 Research.

*************

HPE and AMD are partnering to deliver the next generation of AI and supercomputing infrastructure. Combining HPE’s deep expertise in large-scale system design with AMD EPYC™ processors and AMD Instinct™ accelerators, the two companies help customers build platforms capable of supporting the most demanding AI, modeling, simulation, and analytics workloads. Whether accelerating scientific research, enabling digital engineering, or training advanced AI models, HPE and AMD focus on what matters most: faster insights, greater efficiency, and the flexibility to scale. Together, they provide the technology foundation organizations need to solve complex problems, drive discovery, and create lasting business and societal impact.