WhiteFiber launched Continuum, a cross-data-center networking architecture designed to eliminate bandwidth constraints in distributed GPU supercluster operations.
WhiteFiber, Inc., a provider of AI infrastructure solutions, has announced the commercial availability of WhiteFiber Continuum, a cross-data-center networking solution and the first commercially available distributed GPU supercluster architecture. WhiteFiber Continuum links geographically separated data centers into a single logical GPU supercluster, removing bandwidth and throughput as constraints on how and where AI infrastructure can be deployed.
The system is built on two HITRUST-certified QTS facilities located 83 kilometers apart, joined into one logical system over 12 Zayo dark fiber strands and powered by DriveNets AI Fabric and WEKA NeuralMesh AI data storage and memory infrastructure. The architecture delivers 136 Tbps of bandwidth at 0.9 milliseconds of guaranteed round-trip latency. WhiteFiber has submitted patent applications for the underlying implementation.
"WhiteFiber Continuum is the infrastructure answer to a problem the industry has lived with for years: geographic distance as a ceiling on what a cluster can do," said Sam Tabar, Chief Executive Officer of WhiteFiber. "Today we are making it commercially available to enterprises that need AI compute to perform at scale, hold up under compliance requirements, and not break when a single site has a problem. This is the foundation for a new class of AI infrastructure, and we are ready to build it with our customers."
Many telecom and metro facilities possess power and fiber capacity suitable for AI workloads. WhiteFiber Continuum allows these sites to contribute to a single logical GPU supercluster rather than operate as isolated, underutilized capacity—turning stranded telco and metro assets into infrastructure that can support last-mile inference and agent workloads. For enterprises, the same architecture enables pooling GPU resources across sites to overcome the power and space constraints of a single campus, adding overflow training and inference capacity without requiring a separate greenfield deployment.
For regulated industries, distributing a cluster across sites introduces questions of data residency. Continuum is engineered to keep sensitive workloads within their originating jurisdiction even while failover and pooled AI compute occur across locations, allowing enterprises to gain resilience without sacrificing data sovereignty. Because no single site represents a single point of failure, the system is designed so that training runs can continue when one location goes offline, rather than stalling until recovery.
WhiteFiber Continuum was developed in collaboration with DriveNets, which provides the high-performance AI network fabric connecting both sites, and WEKA, which provides high-performance data storage and memory infrastructure across the cluster. "WhiteFiber's Continuum shows what becomes possible when ambitious engineering is paired with the right networking architecture," said Yossi Kikozashvili, VP Product and GTM – AI Infra, at DriveNets. "DriveNets' Ethernet-based AI fabric delivers the highest performance even in the most demanding, high-bandwidth, low-latency environments, ensuring scale-across superclusters move data efficiently, maximize GPU utilization, and optimize power efficiency."
"The industry has treated distance as a hard ceiling on AI infrastructure, and WhiteFiber Continuum breaks it. But a supercluster is only as unified as its data. WEKA's NeuralMesh gives Continuum a single, high-performance storage and memory foundation across sites, so GPUs that sit many kilometers apart operate as if the data were local. This is what turns two data centers into one engine, and what the next generation of AI innovation will be built on," said Liran Zvibel, co-founder and CEO at WEKA.
WhiteFiber Continuum is commercially available as of September 23, 2026. Customers interested in deployment can visit whitefiber.com/whitefiber-continuum.