DriveNets announced that WhiteFiber has implemented its Network Cloud-AI solution to power the back-end and storage network for its cloud-based GPU-as-a-Service (GPUaaS) platform. This deployment enables WhiteFiber to deliver high-performance AI infrastructure with optimal GPU efficiency and cost-effectiveness. The decision was driven by DriveNets’ low latency, scalable architecture, multi-tenant support, and rapid, low-maintenance deployment.
WhiteFiber—an innovative, fast-growing provider of GPUaaS solutions—delivers high-performance infrastructure with no bottlenecks, enabling its customers to benefit from fast deployment, high flexibility and customization, reliability, and affordability. The company uses DriveNets Network Cloud-AI for its back-end network—connecting GPUs to GPUs—and for storage networking—connecting storage to GPUs—in its new AI data center in Iceland.
“WhiteFiber is focused on customer satisfaction by delivering high performance and maximum value through superior GPU utilization,” said Tom Sanfilippo, CTO, WhiteFiber. “We selected DriveNets Network Cloud-AI since it has been proven to deliver the highest Ethernet-based AI connectivity in enterprise and Hyperscaler environments. That’s the type of innovation and performance we look for in the companies we work with. We were able to deploy Network Cloud-AI with very short lead time and exceptionally fast installation time, getting our AI data center up and running quickly, which was critical to address rapidly growing market demand.”
Based on DriveNets Network Cloud-AI infrastructure, White Fiber has seen higher collective communications performance (NCCL testing bus bandwidth), with a significant improvement over other Ethernet solutions tested, yielding better job completion time for their customers’ workloads.
Multi-Tenancy Supporting GPUaaS Requirements
DriveNets’ multi-tenancy support, which dynamically adapts to any model, is specifically critical to ensure the performance of NeoClouds’ GPUaaS offerings.
“DriveNets’ AI solution can uniquely address the needs of NeoCloud (GPUaaS) providers that require support for multi-tenancy,” said Ido Susan, DriveNets CEO and Founder. “By using a scheduled fabric rather than endpoint-based scheduling, we can enforce the separation between different tenants’ traffic and resources better than other solutions, and eliminate the ‘noisy neighbor effect’ that impacts the performance of multiple workloads running over the same cluster.”
“The match between DriveNets and WhiteFiber is not a coincidence – both companies are innovative and committed to delivering the highest performance and scale to their customers, with a fast delivery experience,” said Alan Weckel, Founder and Technology Analyst at 650 Group. “WhiteFiber’s selection of DriveNets supports the clear trend that we are seeing in the market – the move to Ethernet-based AI connectivity for greater vendor flexibility.”
Learn more about WhiteFiber and DriveNets network cloud-ai, visit the website here.
Related News:
