Lattice Collaborates with Arm to Advance Secure, Adaptable AI Data Center Infrastructure

Estimated reading time: 1 minute

Lattice Semiconductor, the low power programmable and platform firmware leader, announced a collaboration with Arm to enable secure control, management, and platform adaptability for AI infrastructure. Combining Arm® AGI CPU server platforms with Lattice FPGAs and AMI firmware creates a solution that provides advanced security, manageability, connectivity, and flexibility for next-generation AI data center infrastructure.

In Arm AGI CPU server platforms, Lattice FPGAs work alongside the CPU and baseboard management controller, with AMI firmware providing platform boot and management capabilities. The Lattice FPGAs act as companion chips providing secure control and management capabilities, enabling platform trust, connectivity, and infrastructure adaptability. Lattice FPGA technology also enables Open Compute Project (OCP)-compliant connectivity between the Host Processor Module (HPM) and Datacenter Secure Control Module (DC-SCM) through the LVDS Tunneling Protocol and Interface (LTPI), an open, standardized interface that simplifies platform integration, enhances interoperability, and supports scalable, modular AI infrastructure.

“AI infrastructure is evolving rapidly, creating greater demand for secure, interoperable control and management at every level of the system,” said Mamta Gupta, VP, Business Development, Compute & Communications, Lattice Semiconductor. “Together with Arm, we're helping customers accelerate deployment of next-generation computing platforms through secure management, open standards, and a scalable foundation that can evolve with changing customer needs.

“Scaling AI infrastructure requires an ecosystem working across compute, connectivity, management, and security,” said Eddie Ramirez, Vice President of Go-to-Market, Cloud AI, Arm. “By working with Lattice to integrate secure management and control, we’re expanding the ecosystem around Arm AGI CPU and giving customers more flexibility to build and deploy next-generation AI infrastructure.”