.webp)
Hewlett Packard Enterprise (HPE) and NVIDIA have introduced the NVIDIA AI Computing by HPE portfolio, a suite of AI solutions designed to facilitate the adoption of generative AI within enterprises. Among the key offerings is HPE Private Cloud AI, which integrates NVIDIA's AI computing, networking, and software with HPE's storage, compute, and the HPE GreenLake cloud. This collaboration aims to provide a sustainable, energy-efficient, and flexible platform for developing and deploying generative AI applications.
During the HPE Discover 2024 keynote in Sydney, HPE President and CEO Antonio Neri and NVIDIA founder and CEO Jensen Huang announced the collaboration. Neri emphasised the need to overcome the complexities and risks associated with fragmented AI technology in order to harness the transformative potential of generative AI for enterprises. "To unleash the immense potential of generative AI in the enterprise, HPE and NVIDIA co-developed a turnkey private cloud for AI that will enable enterprises to focus their resources on developing new AI use cases that can boost productivity and unlock new revenue streams," Neri said. Huang added, "Never before have NVIDIA and HPE integrated our technologies so deeply...to equip enterprise clients and AI professionals with the most advanced computing infrastructure and services to expand the frontier of AI."
The HPE Private Cloud AI service offers a cloud-based experience with full lifecycle management, supported by the new OpsRamp AI copilot to improve IT operations. This solution is available in four configurations to support various AI workloads, including those reliant on proprietary data. It incorporates NVIDIA AI Enterprise software, NIM inference microservices, NVIDIA Spectrum-X Ethernet networking, HPE GreenLake for File Storage, and HPE ProLiant servers. The stack supports NVIDIA L40S, NVIDIA H100 NVL Tensor Core GPUs, and the NVIDIA GH200 NVL2 platform.
HPE GreenLake cloud services enhance the cloud experience by providing a single platform-based control plane for managing endpoints, workloads, and data across hybrid environments. HPE's integration of OpsRamp AI infrastructure observability provides insights for managing the end-to-end NVIDIA accelerated computing stack. This includes NVIDIA NIM and AI software, NVIDIA Tensor Core GPUs, and NVIDIA Quantum InfiniBand and Spectrum Ethernet switches.
HPE and NVIDIA have also collaborated closely with global system integrators like Deloitte, HCLTech, Infosys, TCS, and Wipro to ensure rapid value realisation from AI applications. These integrators can assist enterprises in running complex AI workloads across various industries. Huang noted that generative AI and accelerated computing are driving significant transformations across industries as they strive to join the industrial revolution.
Furthermore, the HPE Cray XD670 server, which supports eight NVIDIA H200 NVL Tensor Core GPUs, is ideal for building large language models (LLMs). The HPE ProLiant DL384 Gen12 server, equipped with the NVIDIA GH200 NVL2 platform, is tailored for LLM consumers using larger models or retrieval-augmented generation (RAG) technology. Additionally, the HPE ProLiant DL380a Gen12 server supports up to eight NVIDIA H200 NVL Tensor Core GPUs, providing flexibility for scaling generative AI workloads.
HPE GreenLake for File Storage has achieved certification for NVIDIA DGX BasePOD and validation for NVIDIA OVX systems, offering an enterprise file storage solution tailored for accelerating AI and GPU-intensive workloads at scale. HPE intends to be a time-to-market partner on upcoming NVIDIA reference architecture storage certification programmes.
The joint initiative underscores HPE and NVIDIA's commitment to accelerating industrial AI adoption by providing comprehensive, optimised solutions aimed at simplifying the deployment and management of AI workloads while addressing data privacy, security, and governance requirements.
