Hewlett Packard and Nvidia Launch Private Cloud Solutions for Secure Generative AI Model Deployment

Hewlett Packard Enterprise (HPE) has joined forces with Nvidia to introduce an innovative platform designed for securely running generative AI models and workloads in the cloud. Unveiled during HPE’s Discover event, HPE Private Cloud AI seamlessly integrates Nvidia’s cutting-edge computing and software technologies with HPE’s advanced storage solutions. This powerful combination enables enterprises to execute generative AI tasks efficiently while maintaining robust security measures.

With HPE Private Cloud AI, organizations can perform inference and fine-tuning activities utilizing proprietary data, paving the way for the creation and implementation of flexible and secure generative AI solutions. Jensen Huang, founder and CEO of Nvidia, emphasized the significance of this collaboration, stating, “Generative AI and accelerated computing are driving a profound transformation as every industry strives to participate in this new industrial revolution.” He highlighted the unprecedented depth of technology integration between Nvidia and HPE, showcasing a complete AI computing stack that equips businesses with the most sophisticated infrastructure and services to push the boundaries of AI.

The platform features Nvidia AI Enterprise software, which provides essential tools to facilitate the development of generative AI applications. Notably, the recently introduced NIM microservice enhances the optimization of AI model inference, enhancing the overall user experience. Additionally, HPE’s software suite includes capabilities to monitor AI models and ensure data compliance, addressing critical concerns for enterprises navigating the complexities of AI adoption.

HPE GreenLake, a versatile cloud service, empowers users to automate and manage endpoints and data across hybrid environments effortlessly. The integrated AI infrastructure stack within HPE Private Cloud AI allows enterprise users to efficiently manage and deploy AI workloads across multiple platforms, including HPE GreenLake for File Storage, Nvidia’s Spectrum-X Ethernet networking, and high-performance GPUs such as Nvidia’s H100 Tensor Cores. This strategic alignment facilitates the seamless scaling of generative AI solutions, catering to a growing demand in various industries.

The HPE Private Cloud AI platform is set to become generally available in fall 2024, offering organizations a comprehensive tool for AI initiatives. Furthermore, HPE announced additional developments during the Discover 2024 event, including the extended support for several HPE supercomputing and server solutions by Nvidia. A notable inclusion is the Cray XD670, now enhanced to support eight Nvidia H200 Tensor Core GPUs, making it an exceptional computing resource for enterprises aiming to harness the power of large language models.

Antonio Neri, HPE’s president and CEO, underscored the transformative potential of generative AI for businesses. He noted the challenges posed by fragmented AI technology, which can impede large-scale adoption and risk the protection of proprietary data. To effectively unleash the capabilities of generative AI within enterprises, HPE and Nvidia have collaboratively developed a turnkey private cloud solution. This initiative enables organizations to direct their resources toward creating innovative AI applications that enhance productivity and unlock new revenue opportunities, ensuring they remain at the forefront of this technological revolution.