The Revolutionary Supercomputing Solution for Generative AI

Explore Hewlett Packard Enterprise’s innovative supercomputing solution tailored for generative AI, empowering enterprises with cutting-edge technology for accelerated model training and optimal outcomes.

Hewlett Packard Enterprise has introduced a cutting-edge solution for generative AI, tailored for large enterprises, research institutions, and government bodies. This offering aims to expedite the training and optimization of AI models by leveraging private datasets. The comprehensive solution comprises a software suite facilitating model training and refinement, along with the provision of liquid-cooled supercomputers, accelerated compute capabilities, networking infrastructure, storage, and associated services. These components collectively expedite the realization of AI value.

Justin Hotard, Executive Vice President and General Manager of HPC, AI & Labs at Hewlett Packard Enterprise, highlighted the necessity for purpose-built solutions to effectively and efficiently drive innovation through AI model training. Emphasizing the need for sustainable solutions delivering dedicated performance and scalability akin to a supercomputer, Hotard expressed enthusiasm about the collaboration with NVIDIA. Together, they aim to offer an all-inclusive AI-native solution, significantly accelerating AI model training and enhancing outcomes for customers.

At the heart of this supercomputing solution lies a suite of software tools essential for building AI applications, customizing pre-existing models, and coding development and modification. Integrated with HPE Cray supercomputing technology, powered by NVIDIA Grace Hopper GH200 Superchips, this solution empowers organizations with unparalleled scale and performance crucial for handling extensive AI workloads, such as training large language models and deep learning recommendation models.

This formidable system demonstrated remarkable capabilities, fine-tuning the open-source 70 billion-parameter Llama 2 model in under 3 minutes using the HPE Machine Learning Development Environment, directly translating to faster time-to-value for customers. Leveraging NVIDIA technology alongside HPE’s advanced supercomputing capabilities resulted in a 2-3X improvement in system performance.

Ian Buck, Vice President of Hyperscale and HPC at NVIDIA, highlighted the transformative impact of generative AI across industries and praised the collaboration with HPE. The joint effort aims to provide customers with the necessary performance to drive breakthroughs in their generative AI initiatives.

The supercomputing solution for generative AI is an integrated, purpose-built offering encompassing AI/ML acceleration software, scalability features leveraging HPE Cray EX2500, high-performance networking through HPE Slingshot Interconnect, and turnkey simplicity facilitated by HPE Complete Care Services. This comprehensive suite aims to simplify AI adoption and implementation for organizations.

Recognizing the increasing demand for energy-efficient solutions due to the burgeoning growth of AI workloads, HPE focuses on delivering computing initiatives with liquid-cooling capabilities. These innovations promise up to 20% performance improvement per kilowatt over air-cooled solutions and consume 15% less power, contributing to minimizing the carbon footprint. The supercomputing solution for generative AI integrates direct liquid cooling (DLC) to efficiently cool systems while reducing energy consumption for compute-intensive applications.

HPE stands as a unique facilitator, poised to empower organizations with powerful compute technology while aiding in reducing energy usage—a step toward a sustainable future for supercomputing and AI.

In the rapidly evolving landscape of AI and supercomputing, Hewlett Packard Enterprise’s groundbreaking solution for generative AI stands as a testament to innovation and progress. With an integrated approach that marries cutting-edge technology, scalable infrastructure, and a commitment to energy efficiency, HPE is not just empowering organizations to harness the power of AI but is also paving the way for a more sustainable future. As AI continues to drive transformative change across industries, HPE’s dedication to delivering high-performance solutions while minimizing environmental impact positions it as a key player in shaping the intersection of technology, sustainability, and the next frontier of AI advancements.