d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale Raptor XPU Deployment
Original title:d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment
AI inference chipmaker d-Matrix has adopted NVIDIA NVLink Fusion to integrate its next-generation Raptor XPUs directly into NVIDIA's MGX rack architecture and Spectrum-X networking. This integration allows custom silicon makers to bypass the friction and capital costs of developing proprietary multi-node clustering and cooling systems. By extending NVLink Fusion to third-party processors, NVIDIA is repositioning its interconnect and rack fabric as the foundational substrate for modern AI factories.
Why it's worth reading
It highlights NVIDIA's strategy to turn its proprietary NVLink and MGX rack fabric into an open interconnect standard, absorbing specialized third-party inference silicon.