In an exciting development for cloud computing and artificial intelligence, NVIDIA and Amazon Web Services (AWS) have announced a series of groundbreaking innovations at the AWS re:Invent conference in Las Vegas. This annual event is a gathering point for the global cloud computing community, featuring keynotes and over 2,000 technical sessions. The focus this year is on accelerating AI advancements, enhancing robotics, and simplifying quantum computing research. Among the key announcements are the availability of NVIDIA DGX Cloud on AWS and a suite of enhanced tools for AI, quantum computing, and robotics.
### NVIDIA DGX Cloud on AWS: Empowering AI at Scale
NVIDIA’s DGX Cloud AI computing platform is now available through AWS Marketplace Private Offers. This platform provides a high-performance, fully managed solution that enables enterprises to train and customize AI models effectively. DGX Cloud offers flexible terms, a fully managed and optimized platform, and direct access to NVIDIA experts, thus helping businesses to scale their AI capabilities swiftly. Early adopter Leonardo.ai, which is part of the Canva family, has already begun leveraging DGX Cloud on AWS to develop cutting-edge design tools.
### AWS Liquid-Cooled Data Centers with NVIDIA Blackwell
A significant innovation in AI server technology is the use of liquid cooling, which allows for more efficient cooling of high-density compute chips, leading to better performance and energy efficiency. AWS has developed solutions that integrate configurable liquid-to-chip cooling across its data centers, enhancing both performance and sustainability. This new cooling solution integrates seamlessly with air- and liquid-cooling capabilities for high-performance AI supercomputing systems, such as the NVIDIA GB200 NVL72. This system also supports AWS’s network switches and storage servers. Notably, this flexible, multimodal cooling design will be used for the next-generation NVIDIA Blackwell platform, which underpins Amazon EC2 P6 instances, DGX Cloud on AWS, and Project Ceiba.
### Advancing Physical AI: Accelerated Robotics Simulation with NVIDIA on AWS
NVIDIA is expanding the reach of its Omniverse platform on AWS by introducing NVIDIA Isaac Sim, now running on high-performance Amazon EC2 G6e instances accelerated by NVIDIA L40S GPUs. This reference application allows developers to simulate and test AI-driven robots in realistic virtual environments. One of the key workflows enabled by Isaac Sim is synthetic data generation, which is now further accelerated with the integration of OpenUSD NIM microservices, covering everything from scene creation to data augmentation. Robotics companies such as Aescape, Cohesive Robotics, Cobot, Field AI, Standard Bots, Swiss Mile, and Vention are already using Isaac Sim to simulate and validate their robots’ performance before deployment. Furthermore, companies like Rendered.ai, SoftServe, and Tata Consultancy Services are utilizing synthetic data generation capabilities to bootstrap perception AI models for various robotics applications.
### NVIDIA BioNeMo on AWS: Pioneering AI-Based Drug Discovery
NVIDIA’s BioNeMo NIM microservices and AI Blueprints, designed to advance drug discovery, are now integrated into AWS HealthOmics. This fully managed biological data compute and storage service is tailored to accelerate scientific breakthroughs in clinical diagnostics and drug discovery. This collaboration provides researchers with access to AI models and scalable cloud infrastructure for drug discovery workflows. Biotech companies have already begun using NVIDIA BioNeMo on AWS to enhance their research and development pipelines. For instance, A-Alpha Bio, a Seattle-based biotechnology company, recently published a study detailing a collaborative effort with NVIDIA and AWS to develop an antibody AI model called AlphaBind. This model, running on Amazon EC2 P5 instances with NVIDIA H100 Tensor Core GPUs, achieved a 12x increase in inference speed and processed over 108 million inference calls in two months. Additionally, SoftServe has launched a generative AI solution for drug discovery, built with NVIDIA Blueprints, that promises to deliver faster workflows and will soon be available in AWS Marketplace.
### Real-Time AI Blueprints: Streamlined Deployment for Video, Cybersecurity, and More
NVIDIA’s latest AI Blueprints are now available for immediate deployment on AWS, providing real-time applications such as vulnerability analysis for container security and video search and summarization agents. Developers can integrate these blueprints into existing workflows to accelerate deployments. For instance, the NVIDIA AI Blueprint for video search and summarization enables the creation of visual AI agents that can analyze videos in real-time or from archives to answer user queries, generate summaries, and trigger alerts for specific scenarios. AWS and NVIDIA have collaborated to offer a reference architecture that applies the NVIDIA AI Blueprint for vulnerability analysis, enhancing early security patching in continuous integration pipelines on AWS cloud-native services.
### NVIDIA CUDA-Q on Amazon Braket: Simplifying Quantum Computing
NVIDIA CUDA-Q is now integrated with Amazon Braket, streamlining quantum computing development. CUDA-Q users can access Amazon Braket’s quantum processors, while Braket users benefit from CUDA-Q’s GPU-accelerated workflows for development and simulation. The CUDA-Q platform allows developers to build hybrid quantum-classical applications and run them on a variety of quantum processors, both simulated and physical. Preinstalled on Amazon Braket, CUDA-Q offers a seamless development platform for hybrid quantum-classical applications, unlocking new possibilities in quantum research.
### Enterprise Platform Providers and Consulting Leaders Enhance AI with NVIDIA on AWS
Leading software platforms and global system integrators are aiding enterprises in rapidly scaling generative AI applications built with NVIDIA AI on AWS, fostering innovation across multiple industries. Cloudera, for example, is using NVIDIA AI on AWS to boost its new AI inference solution, aiding Mercy Corps in improving the precision and effectiveness of its aid distribution technology. Cohesity has integrated NVIDIA NeMo Retriever microservices into its generative AI-powered conversational search assistant, Cohesity Gaia, to enhance the recall performance of retrieval-augmented generation. Cohesity customers on AWS can leverage this integration within Gaia. DataStax has announced that Wikimedia Deutschland is using the DataStax AI Platform to make Wikidata available to developers as an embedded vectorized database. Built with NVIDIA NeMo Retriever and NIM microservices, this platform is available on AWS. Deloitte’s C-Suite AI now supports NVIDIA AI Enterprise software, including NVIDIA NIM microservices and NVIDIA NeMo, for CFO-specific use cases, such as financial statement analysis, scenario modeling, and market analysis.
### RAPIDS Quick Start Notebooks: Enhancing Data Science on Amazon EMR
NVIDIA and AWS are accelerating data science and data analytics workloads with the RAPIDS Accelerator for Apache Spark, which enhances analytics and machine learning workloads without requiring code changes and can reduce data processing costs by up to 80%. Quick Start notebooks for the RAPIDS Accelerator for Apache Spark are now available on Amazon EMR, Amazon EC2, and Amazon EMR on EKS. These notebooks provide a simple way to optimize Spark jobs for maximum performance on GPUs within AWS EMR.
### Powering Next-Generation Industrial Edge Systems with NVIDIA and AWS
The NVIDIA IGX Orin and Jetson Orin platforms now integrate seamlessly with AWS IoT Greengrass, facilitating the deployment and operation of AI models at the edge and managing fleets of connected devices efficiently. This integration enhances scalability and simplifies the deployment process for industrial and robotics applications. Developers can utilize NVIDIA’s advanced edge computing power alongside AWS’s purpose-built IoT services, creating a secure and scalable environment for autonomous machines and smart sensors. A guide for getting started, authored by AWS, is now available to support developers in utilizing these capabilities.
These announcements underscore NVIDIA’s commitment to advancing enterprise-ready industrial edge systems, enabling rapid and intelligent operations in real-world applications. Attendees at AWS: re:Invent 2024 can explore NVIDIA’s innovations through live demos, technical sessions, and hands-on labs, gaining deeper insights into the future of cloud computing and AI. For more detailed information, visit the official websites of NVIDIA and AWS.
For more Information, Refer to this article.
