WhiteFiber launches Continuum: 83 km two-site GPU supercluster at 136 Tbps

WhiteFiber has made Continuum commercially available, turning its Project Redwood two-site GPU supercluster into a deployable product. It links two HITRUST-certified QTS data centers 83 km apart over 12 Zayo dark fiber strands, delivering 136 Tbps aggregate bandwidth and 0.9 ms guaranteed round-trip latency, with DriveNets AI Fabric and WEKA NeuralMesh.

• Two QTS sites 83 km apart linked by 12 Zayo dark fiber strands
• 136 Tbps aggregate: 170 800G channels, 0.9 ms latency
• July R&D test measured 111.2 Tbps on part of the spectrum
• Patent applications filed for the underlying implementation