Fortinet announced an integrated security solution that runs FortiGate VM directly on NVIDIA BlueField-3 data processing units, embedding firewalling and policy enforcement into accelerated AI data center infrastructure. The collaboration targets AI factories and private cloud environments where traditional host-based security architectures struggle to keep pace with high-throughput, low-latency workloads.
By moving core security functions off the host CPU and onto the BlueField-3 DPU, Fortinet and NVIDIA place firewalling, segmentation, and zero-trust controls directly within the infrastructure fabric. The approach enables security inspection and enforcement to operate independently of GPU and CPU workloads, improving performance isolation while maintaining consistent policy enforcement across multitenant AI environments.
The solution supports scalable architectures for enterprises, cloud service providers, and telco edge deployments. Running on FortiOS, FortiGate VM delivers centralized management and visibility while enforcing segmentation policies inside the network fabric itself, aligning security operations with emerging AI-driven data center designs that rely on accelerated networking and disaggregated infrastructure.
- FortiGate VM runs natively on NVIDIA BlueField-3 DPUs, offloading security from host CPUs
- Embedded firewalling, segmentation, and zero-trust enforcement inside the infrastructure fabric
- Improved isolation and multitenancy for AI factories and private cloud environments
- Reduced latency and higher throughput by bypassing host-based security processing
- Validated deployment using Open vSwitch and VXLAN tunnels for WAN and LAN integration
“By moving firewalling, segmentation, and zero-trust controls on the DPU, we help organizations improve isolation, reduce latency, and simplify consistent policy enforcement across their environments,” said John Whittle, Chief Operating Officer at Fortinet.
🌐 Analysis
The Fortinet–NVIDIA collaboration reflects a broader shift toward infrastructure-embedded security as AI data centers adopt DPU-accelerated networking to preserve GPU efficiency. As hyperscalers and enterprises design AI factories around offloaded networking and storage, integrating security at the DPU layer aligns protection with the same architectural principles driving performance and scale.
🌐 We’re tracking the latest developments in AI infrastructure and data center networking. Follow our ongoing coverage at: https://convergedigest.com/category/ai-infrastructure/
🌐 We’re launching the “Data Center Networking for AI” series on NextGenInfra.io and inviting companies building real solutions—silicon, optics, fabrics, switches, software, orchestration—to share their views on video and in our expert report. To get involved, send a note to [email protected] or [email protected].
