Advancing Cloud Evolution: AWS and Nvidia Propel AI to the Next Level

The cloud giant has officially introduced a new Amazon EC2 P5 instance, powered by NVIDIA H100 Tensor Core GPUs. This service allows users to effortlessly scale generative AI, high-performance computing (HPC), and other applications with just a simple click from their browser.

This announcement comes at a time when AI is experiencing a significant breakthrough, comparable to the iPhone's impact. Developers and researchers are constantly discovering new applications for AI using large language models (LLMs). To bring these exciting use cases to market efficiently, accelerated computing is essential.

The NVIDIA H100 GPU achieves supercomputing-class performance, thanks to architectural innovations, such as fourth-generation Tensor Cores, a new Transformer Engine for accelerating LLMs, and the latest NVLink technology, enabling GPUs to communicate with each other at an incredible speed of 900GB/sec.

Enabling Scalability with P5 Instances

Amazon EC2 P5 instances present an optimal solution for training and executing inference on increasingly intricate large language models (LLMs) and computer vision models. These neural networks drive the most demanding and compute-intensive generative AI applications, spanning various tasks such as question answering, code generation, video and image synthesis, speech recognition, and more.

P5 instances can be deployed in hyperscale clusters, known as EC2 UltraClusters, consisting of high-performance computing, networking, and storage resources in the cloud. Each EC2 UltraCluster acts as a potent supercomputer, empowering customers to execute their most complex AI training and distributed high-performance computing (HPC) workloads across multiple systems.

To facilitate applications requiring extensive communication between compute nodes at scale, the P5 instance incorporates petabit-scale non-blocking networks, powered by AWS EFA, offering a 3,200 Gbps network interface for Amazon EC2 instances.

With P5 instances, machine learning applications can leverage the NVIDIA Collective Communications Library to utilize up to 20,000 H100 GPUs.

NVIDIA AI Enterprise plays a crucial role in enhancing the capabilities of P5 instances, providing users with a comprehensive suite of software encompassing over 100 frameworks, pretrained models, AI workflows, and tools to optimize AI infrastructure.

Engineered to simplify the development and deployment of AI applications, NVIDIA AI Enterprise addresses the complexities associated with building and maintaining a high-performance, secure, cloud-native AI software platform. Accessible through the AWS Marketplace, it offers continuous security monitoring, regular and timely patching to address common vulnerabilities and exposures, API stability, enterprise support, and direct access to NVIDIA AI experts.

Customer Testimonials

The collaboration between NVIDIA and AWS in bringing GPU acceleration to the cloud for over a decade has been remarkable. The introduction of the new P5 instances is yet another testament to this fruitful partnership, representing a significant advancement in delivering cutting-edge performance that empowers developers to pioneer the next generation of AI.

Here are a few examples of what our customers are already saying:

Anthropic is building dependable, interpretable, and controllable AI systems, presenting numerous opportunities to generate commercial value and benefit the public.

"While the large, general AI systems of today can have significant benefits, they can also be unpredictable, unreliable and opaque, so our goal is to make progress on these issues and deploy systems that people find useful. We expect P5 instances to deliver substantial price-performance benefits over P4d instances, and they’ll be available at the massive scale required for building next-generation LLMs and related products", - said Tom Brown, co-founder of Anthropic.

Cohere, a leading pioneer in language AI, empowers every developer and enterprise to build products with world-leading natural language processing (NLP) technology while keeping their data private and secure.

"Cohere leads the charge in helping every enterprise harness the power of language AI to explore, generate, search for and act upon information in a natural and intuitive manner, deploying across multiple cloud platforms in the data environment that works best for each customer. NVIDIA H100-powered Amazon EC2 P5 instances will unleash the ability of businesses to create, grow and scale faster with its computing power combined with Cohere’s state-of-the-art LLM and generative AI capabilities", - said Aidan Gomez, CEO of Cohere.

For its part, Hugging Face is on a mission to democratize good machine learning.

"As the fastest growing open-source community for machine learning, we now provide over 150,000 pretrained models and 25,000 datasets on our platform for NLP, computer vision, biology, reinforcement learning and more. We’re looking forward to using Amazon EC2 P5 instances via Amazon SageMaker at scale in UltraClusters with EFA to accelerate the delivery of new foundation AI models for everyone", - said Julien Chaumond, chief technology officer and co-founder of Hugging Face.

Today, more than 450 million people around the world use Pinterest as a visual inspiration platform to shop for products personalized to their taste, find ideas and discover inspiring creators.

"We use deep learning extensively across our platform for use cases such as labeling and categorizing billions of photos that are uploaded to our platform, and visual search that provides our users the ability to go from inspiration to action. We’re looking forward to using Amazon EC2 P5 instances featuring NVIDIA H100 GPUs, AWS EFA and UltraClusters to accelerate our product development and bring new empathetic AI-based experiences to our customers", - said David Chaiken, chief architect at Pinterest.

Conclusion

The AWS and Nvidia collaboration has propelled AI to new heights with Amazon EC2 P5 instances, powered by NVIDIA H100 Tensor Core GPUs. These instances enable effortless scaling of generative AI and high-performance computing applications, unlocking novel use cases and accelerating development processes. Incorporating petabit-scale non-blocking networks and NVIDIA Collective Communications Library support, machine learning applications can fully harness the potential of AI. Customer testimonials underscore the significant impact of P5 instances, showcasing enhanced performance and innovative AI-driven experiences. P5 instances mark a crucial step towards democratizing AI and fostering a future of AI-powered solutions for all.