SUSE Unveils Sovereign Control for Enterprise Agentic AI with NVIDIA OpenShell Reference Architecture and SLES for NVIDIA

UAE

DUBAI — Open source infrastructure company SUSE has introduced new enterprise sovereign capabilities designed to give organizations complete operational custody, compliance, and governance over autonomous agentic AI workflows. Announced in Dubai, the company is enhancing enterprise perimeter security for autonomous artificial intelligence agents by embedding the NVIDIA OpenShell Reference Architecturedirectly into SUSE AI Factory with NVIDIA, establishing an audited framework that unites Kubernetes sandboxing, API security gateways, and hardware accelerated policy enforcement.

The combined architecture coordinates NVIDIA Vera Rubin systems for high performance foundation model inference, NVIDIA Vera CPUs executing OpenShell for local agent operations, and NVIDIA BlueField 4 Data Processing Units (DPUs) running DOCA powered NVIDIA Sentry for host independent telemetry inspection and in silicon security enforcement. Through active zero trust governance, supervisor proxies continuously inspect internal agent network calls, authenticate tool access against corporate policies, and block external data egress to protect enterprise intellectual property. In parallel, private model serving executed via NVIDIA NIM microservices removes dependence on third party per token public application programming interfaces, shifting corporate artificial intelligence outlays toward predictable, capacity anchored infrastructure economics.

Addressing enterprise operating systems, SUSE launched SUSE Linux Enterprise Server (SLES) for NVIDIA, engineered to deliver day zero driver and runtime compatibility for bleeding edge hardware including the Vera Rubin architecture. Scheduled for general availability in November 2026 within the SUSE AI Factory stack, the platform bridges the gap between sandbox testing and rack scale data center production by consolidating technical support across the SLES 16 base operating system, the Linux kernel, GPU drivers, and Kubernetes container runtimes under a single unified enterprise service level agreement.

The sovereign operating framework is already seeing active sovereign deployment within the public sector, where Mexico’s Guanajuato Secretariat of Finance is conducting proof of concept validations with SUSE AI Factory with NVIDIA to pioneer cognitive government services while safeguarding citizen privacy and retaining complete administrative control over municipal data registries.

To foster an enterprise grade deployment ecosystem, SUSE established a two tiered independent software vendor (ISV) certification framework comprising Ready for and Certified for designations. The initial cohort of six certified ecosystem technologies spans the Klique enterprise AI control plane by ClearML, high performance key value cache management from Tensormesh, runtime model context protocol gateways from Traefik Labs, AlchemiStudio sovereign governance engines, Wiser Technology’s Preslav digital twin platform, and CloudCasa by Catalogic for Kubernetes vector backup and disaster recovery. Concurrently, SUSE and Fujitsu have entered into a strategic collaboration to deliver an enterprise hybrid quantum computing stack running NVIDIA CUDA Q software on Fujitsu enterprise servers, bringing hybrid quantum simulations out of research laboratories into private enterprise data centers.

Previous
Previous

Singapore and UAE Forge Strategic AI, Energy, and Advanced Tech Alliances as Prime Minister Lawrence Wong Engages Abu Dhabi Leadership

Next
Next

Vodacom Deploys $97.4 Million Network Expansion Across Mpumalanga and KwaZulu-Natal to Accelerate 5G and Energy Resilience