Search
NVIDIA and Oracle have announced the expansion of Oracle Cloud Infrastructure's (OCI) Supercluster to a multi-gigawatt scale based on the NVIDIA Vera Rubin NVL72 system, addressing the growing demand for AI training, inference, and agent-based applications. Currently, the NVIDIA Vera Rubin NVL72 system and NVIDIA Vera CPU are fully operational on OCI. The core of this collaboration lies in providing customers with infrastructure capable of handling two types of workloads simultaneously: GPU-intensive cutting-edge model training and inference, and CPU-intensive tasks executed by agents between model calls. Customers can independently scale CPU and GPU resources, achieving precise alignment of compute power configuration with workloads, thereby reducing idle capacity and enhancing infrastructure utilization efficiency.
-
C114 Communication Network -
Communication Home
