Home / Technology / HP, Red Hat, NVIDIA: AI Goes Local
HP, Red Hat, NVIDIA: AI Goes Local
9 Sep
Summary
- New platform enables AI deployment from data center to edge.
- Combines HP ZGX Fury and Red Hat AI Factory with NVIDIA.
- Focuses on enhancing local AI inference and orchestration.
HP Inc. has announced a significant collaboration with Red Hat and NVIDIA to introduce an open, enterprise-grade AI platform. This initiative aims to provide organizations with greater flexibility in deploying AI workloads, allowing them to operate locally, in the cloud, or across hybrid environments.
The planned solution integrates HP ZGX Fury, featuring NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip, with Red Hat AI Factory. This powerful combination is designed to enhance AI and orchestration capabilities, enabling production inference closer to users and data sources.
Customers will have the opportunity to evaluate the platform within a sandboxed environment on HP devices running Red Hat AI Factory with NVIDIA. This testing phase will precede the migration of use cases into full production, ensuring a smooth transition.
The platform is engineered to maximize local AI inference throughput, potentially reaching up to 20 PFLOPS FP4 AI performance. It also focuses on reducing environment setup time and deployment risks, while optimizing GPU utilization through advanced NVIDIA CUDA libraries and multi-GPU workload orchestration.
