HPE Partners with Nvidia to Offer “Turnkey” Development and Deployment of GenAI

Hewlett Packard Enterprise (HPE) has partnered with Nvidia to offer what they bill as an integrated “turnkey” solution for organizations looking to adopt generative artificial intelligence (GenAI), but are discouraged by the complexity of the development and management of such workloads.

Named Nvidia AI Computing by HPE, the portfolio of products and services includes co-developed AI applications and will see the two companies jointly introduce and deliver solutions to customers. They will do this alongside channel partners such as Deloitte, Infosys and Wipro.

Also read: Impact of AI on employment: 86% of workers fear job losses, but here is some good news

The expansion of the decades-spanning HPE-Nvidia partnership was announced during HPE President and CEO Antonio Neri’s keynote address at HPE Discover 2024 this week at the Sphere in Las Vegas. He was joined on stage by Nvidia founder and CEO Jensen Huang.

Neri emphasized that GenAI holds significant transformative power, but the complexity of fragmented AI technology comes with too many risks that hinder its large-scale adoption by businesses. Rushing to adopt it can be costly, especially for a company’s most valuable asset: its data, he said.

Huang added that there are three key components in AI, namely large language models (LLMs), the computing resources needed to process these models and data. Therefore, businesses will need an IT stack, a model stack, and a data stack. Each of these elements is complex to deploy and manage, he said.

The HPE-Nvidia partnership worked to produce these models, leveraging Nvidia’s AI Enterprise software platform, including Nvidia NIM inference microservices, and HPE AI Essentials software, which provides AI and selected data foundation as well as a centralized control panel.

The “turnkey” solution will allow organizations that do not have the time or expertise to bring together all capabilities, including training models, to focus their resources on developing new use cases for the ‘AI, Neri said.

The key to this is HPE Private Cloud AI, he said, which offers an integrated AI stack including Nvidia Spectrum-X Ethernet networking, HPE GreenLake for file storage and HPE ProLiant servers optimized for support Nvidia’s L40S, H100 NVL Tensor Core GPUs. and GH200 NVL2 platform.

Also: Latest AI training benchmarks show Nvidia has no competition

AI requires a hybrid cloud by design to deliver GenAI efficiently and across the full AI lifecycle, Neri said, echoing what he said in March at Nvidia GTC. “From training and tuning models on-premises, in a colocation facility or in the public cloud, to inference at the edge, AI is a hybrid cloud workload,” he said.

With the HPE-Nvidia integrated offering, Neri allows users to configure their AI deployment in just three clicks and 24 seconds.

Huang said: “GenAI and accelerated computing are fueling a fundamental transformation as every sector strives to join the industrial revolution. Never before have Nvidia and HPE integrated our technologies so deeply, combining the entire Nvidia AI computing stack with HPE’s private cloud technology. “

Remove complexities and disconnect

The joint solution brings together technologies and teams that are not necessarily integrated within organizations, said Joseph Yang, general manager of HPC and AI for HPE’s Asia Pacific and India region.

AI teams (in companies that have them) typically operate independently of IT teams and may not even report to IT, Yang said in an interview with ZDNET on the sidelines of HPE Discover. They know how to build and train AI models, while IT teams are familiar with cloud architectures that host general-purpose workloads and may not understand AI infrastructures.

Also: Generative AI’s biggest challenge is showing ROI – here’s why

There is a gap between the two, he said, emphasizing that AI and cloud infrastructures are markedly different. Cloud workloads, for example, tend to be small, with one server capable of hosting multiple virtual machines. In comparison, AI inference workloads are large and running AI models requires much larger infrastructures, making these architectures complicated to manage.

IT teams are also facing increasing pressure from management to adopt AI, further adding to the pressure and complexity of deploying GenAI, Yang said.

He added that organizations need to decide what architecture they need to advance their AI projects because their existing hardware infrastructure is a mix of servers that may be outdated. And because they may not have invested in a private cloud or server farm to run AI workloads, they face limits in what they can do since their existing environment does not is not scalable.

“Businesses will need suitable IT infrastructure and capabilities that enable them to accelerate innovation while minimizing the complexities and risks associated with GenAI,” Yang said. “The Nvidia AI Computing by HPE portfolio will enable businesses to accelerate time to value with GenAI to drive new opportunities and growth.

Also: AI skills or AI-enhanced skills? What Employers Need Could Depend on You

Neri further noted that private cloud deployment will also address concerns that organizations may have regarding security and data sovereignty.

He added that HPE complies with all local regulations and compliance requirements, so AI principles and policies will be applied based on local market needs.

According to HPE, the private cloud AI offering supports inference, fine-tuning, and RAG (retrieval augmented generation) AI workloads that leverage proprietary data, as well as controls for data privacy, security and compliance. It also offers cloud ITOps and AIOps capabilities.

Powered by HPE GreenLake Cloud Services, the private cloud AI offering will enable businesses to automate and orchestrate endpoints, workloads and data in hybrid environments.

Also: How my 4 favorite AI tools help me be more productive at work

HPE Private Cloud AI is expected to be available in the fall, alongside the HPE ProLiant DL380a Gen12 server with Nvidia H200 NVL Tensor Core GPU and the HPE ProLiant DL384 Gen12 server with dual Nvidia GH200 NVL2.

General availability of the HPE Cray XD670 Server with Nvidia H200 NVL is planned for summer.

Eileen Yu reported for ZDNET at HPE Discover 2024 in Las Vegas, hosted by Hewlett Packard Enterprise.