Unlock GenAI with HPE ProLiant DL384 Gen 12 and NVIDIA® AI Computing by HPE portfolio

[ComputeExperts]

Across all industries, organizations are leveraging artificial intelligence (AI), particularly large language models (LLMs), to power new generative AI (GenAI) applications such as text generation, language translation, coding, visual content, drug discovery, and many more.

The generative AI industrial revolution has begun

As the scale of data increases, an AI model’s ability to learn and generate more accurate and diverse responses can improve. More data, however, places greater demands on computational resources. To meet the ever-growing demands for resources, traditional data centers need a simpler approach to scaling and integrating accelerated compute in the data center.

Now is the time to embrace enterprise AI

In today’s hybrid reality, where an increasing number of processes are AI-supported and data-driven, organizations need to embrace enterprise AI—where AI is operationalized across the organization to solve a multitude of challenges and drive new innovations. To support these ever-increasing demands, AI technologies need to be developed, deployed, and managed at scale. A critical success factor of embracing enterprise AI is ensuring data center infrastructure is prepared to support this technology shift.

To help enterprises unlock scale-out accelerated computing for GenAI, HPE and NVIDIA® deliver HPE ProLiant Compute DL384 Gen12 with NVIDIA GH200 NVL2, part of the NVIDIA AI Computing by HPE portfolio. This next-generation 2P server provides next-level performance for enterprise AI—enabling a new era of AI.

This versatile system enables enterprises to:

• Accelerate the shift to generative AI. Your organization can leverage artificial intelligence (AI), particularly large language models (LLMs), for AI fine-tuning and inference with Retrieval Augmented Generation (RAG). You can enable new GenAI applications such as text generation, language translation, coding, visual content, and many more.
• Maximize data center utilization. NVIDIA dual GH200 NVL2 with 1.2 terabytes of fast, unified, and coherent memory supports mixed and memory-intensive workloads for next-level performance and maximizes data center utilization for AI computing tasks.
• Get scale-out accelerated computing and enterprise AI productivity. Designed to deploy large language models for AI fine-tuning and inference with RAG with 3.5x capacity and 2X higher performance[1], this versatile scale-out platform significantly enhances computing capabilities.  For faster enterprise AI deployment and success, you can leverage HPE Private Cloud AI.

Ready for next-level AI performance?

Contact your HPE representative today to learn how HPE ProLiant Compute DL384 Gen12 with NVIDIA dual GH200 NVL2 can help you:

• Boost performance per GPU with 1.2 TB coherent memory.
• Increase performance for AI and other mixed workloads, such as job scheduling seismic imaging or financial modelling across systems and GPUs.
• Optimize bandwidth from CPU to GPU to handle demanding AI workloads such as large-scale simulation, weather forecasting, and more.
• Increase flexibility to tackle the challenges of AI, model fine tuning and inference with a RAG vector database in system RAM.
• Create a versatile, scale-out, accelerated computing platform to power the latest LLMs.
• Work with HPE AI experts to build and deploy a custom-tailored AI solution.
• Simplify management of your IT infrastructure by choosing flexible, scalable HPE GreenLake Flex Solutions.

Accelerate your path to production!

Accelerate your path to production AI with HPE Private Cloud AI, a turnkey full stack private cloud. Part of the NVIDIA AI Computing by HPE portfolio, this co-developed scalable, pre-configured, AI-ready private cloud gives AI and IT teams powerful tools to innovate while simplifying ops and keeping your data under your control. HPE Private Cloud AI, a first-of-its-kind solution provides the deepest integration to date of NVIDIA AI computing, networking and software with HPE’s AI storage, compute and the HPE GreenLake cloud. The HPE ProLiant Compute DL384 Gen12 is included in one of four right-sized configurations, enabling enterprises of every size to gain an energy-efficient, fast, and flexible path for sustainably developing and deploying generative AI applications.

Available fall of 2024

HPE ProLiant DL384 Gen12 server with dual NVIDIA GH200 NVL2 is expected to be generally available in the fall of 2024. To learn more about this new compute platform, please visit our website.

Meet HPE blogger, Greg Schmidt, Product Manager for the exciting new Grace Hopper ProLiant server

Greg has experience in HPE from the field to the boardroom. Greg kicked off the HPC and AI Ambassador program, designed successful HPC and AI clusters for customers as a Solution Architect, managed the HPE Apollo team, and led the HPE Apollo 6500 GPU server for multiple years. 

With an extensive background in Graphics Processing Units (GPU), deep learning, virtualization technologies, marketing, sales, product development, and global value chain development, Greg has hands-on business development experience in both global and regional markets.

Greg holds a master’s degree in physics, a Green Belt in Lean Six Sigma, and has been cross-trained in high performance networking, storage, project management, business strategy, and other disciplines. 

A favorite quote:

Always ask your customers “Why,” then listen closely.  Craig Yamasaki

[1] Compared to NVIDIA H100 accelerators