Inside NVIDIA CEO's Speech at Microsoft Ignite 2023

Table of Contents

1. Introduction
2. Accelerated Computing: A Full Stack Challenge
   • 2.1 Transformations in the Computer Industry
   • 2.2 Initiatives in Accelerated Computing
3. The World's Fastest AI Supercomputer
   • 3.1 Building the AI Supercomputer
   • 3.2 Implications and Achievements
4. Unlocking Powerful AI Machines with Windows
   • 4.1 Collaboration and Partnership with Microsoft
   • 4.2 Making Windows a First-Class Client for Large Language Models
5. Introducing NVIDIA Omniverse and Omniverse Cloud
   • 5.1 Industrial Digitalization with Omniverse
   • 5.2 Availability of Omniverse on Azure Cloud
6. AI Foundry Service: Enabling Enterprise AI
   • 6.1 The Opportunity of Generative AI
   • 6.2 Three Ways Companies can Utilize Generative AI
7. NVIDIA's Three Pillars for Custom Large Language Models
   • 7.1 Leveraging AI Expertise, AI Enterprise, and AI Factories
   • 7.2 Building Proprietary Language Models with NVIDIA
8. The Arc of AI Innovation
   • 8.1 Generative AI as a Platform Transition
   • 8.2 Significance of AI Factories and Co-Pilots
   • 8.3 Three Waves of AI Innovation
9. The Microsoft-NVIDIA Partnership
   • 9.1 Transforming Industries through Collaboration
   • 9.2 Advancing Computer Science and Inventing New Technologies

Introduction

The collaboration between two teams, NVIDIA and Microsoft, has resulted in an incredible amount of progress in the computer industry over the past 12 months. This partnership has been focused on accelerated computing, which is a full stack challenge encompassing various aspects from computing to networking. The transformation brought about by generative AI has allowed the teams to achieve significant breakthroughs and deploy large language models on a massive Scale. In this article, we will explore the accomplishments of the NVIDIA-Microsoft collaboration and the impact it has had on the field of AI and computing. We will Delve into topics such as the development of the world's fastest AI supercomputer, the integration of AI capabilities into Windows, the introduction of NVIDIA Omniverse and Omniverse Cloud, and the creation of an AI Foundry Service. Additionally, we will discuss the three pillars that NVIDIA offers for building custom large language models and the vision for the future of AI innovation. Throughout this article, we will highlight the key milestones, benefits, and implications of the NVIDIA-Microsoft partnership, as well as delve into the significance of the collaboration for the broader technology ecosystem.

Accelerated Computing: A Full Stack Challenge

2.1 Transformations in the Computer Industry

In the last 12 months, both the NVIDIA and Microsoft teams have been working tirelessly on accelerated computing, which has brought about significant transformations in the computer industry. This full stack challenge encompasses various aspects, ranging from computing to networking. The progress made during this period is unprecedented, surpassing achievements in any other time frame. The advancements in generative AI have played a crucial role in driving these transformations and revolutionizing the way computers operate. The impact of generative AI has been felt on a global scale, resulting in improved energy efficiency, reduced carbon footprint, cost savings for customers, and enhanced overall performance.

2.2 Initiatives in Accelerated Computing

An essential initiative undertaken by the NVIDIA and Microsoft teams is accelerated computing, aimed at offloading general-purpose computing to accelerate software performance. By leveraging the power of AI, this initiative has enabled significant improvements in energy efficiency, reducing costs for customers while delivering superior performance. To showcase the progress made in accelerated computing, the teams collaborated on building the world's fastest AI supercomputer. In a remarkable feat, this supercomputer was planned and implemented within a few months, making it the fastest AI supercomputer globally. Furthermore, it stands as the third fastest supercomputer overall, highlighting the extraordinary achievements made possible through accelerated computing.

The World's Fastest AI Supercomputer

3.1 Building the AI Supercomputer

The collaboration between NVIDIA and Microsoft has resulted in the creation of an AI supercomputer that holds the title of being the fastest in the world. Traditionally, it would take several years to plan and set up such an advanced computing system. However, the combined efforts of the NVIDIA and Microsoft teams allowed them to achieve this feat within a remarkably short time frame. In just a few months, the AI supercomputer was successfully installed and operational, marking a significant milestone in the field of AI and computing. The speed and efficiency with which this AI supercomputer was built demonstrate the capabilities of accelerated computing and the power of collaboration.

3.2 Implications and Achievements

The world's fastest AI supercomputer represents a monumental achievement for both NVIDIA and Microsoft. The implications of this accomplishment extend beyond mere speed. It solidifies the position of both companies as leaders in the field of AI and sets new standards for performance. By leveraging the capabilities of this supercomputer, a wide range of AI applications can be developed and executed at an unprecedented pace. The availability of such advanced AI technology allows businesses and researchers to push the boundaries of innovation and make significant strides in various domains, including healthcare, finance, and scientific research.

Unlocking Powerful AI Machines with Windows

4.1 Collaboration and Partnership with Microsoft

Microsoft's collaboration with NVIDIA goes beyond building AI supercomputers. One of the most significant aspects of this partnership is the integration of AI capabilities into Windows, making it a first-class client for large language models. Through a collaborative and partner-oriented approach, Microsoft has embraced NVIDIA's platforms and ecosystem, allowing it to host NVIDIA's software stacks on Azure. This integration has opened up new opportunities for developers and users alike, enabling them to harness the power of AI in a seamless and efficient manner.

4.2 Making Windows a First-Class Client for Large Language Models

The joint efforts of Microsoft and NVIDIA have resulted in a groundbreaking development – Windows as a first-class client for large language models. By incorporating NVIDIA's AI technology, Windows has become a powerful platform for utilizing large language models. This integration has led to the widespread adoption of NVIDIA GPUs, making Windows the largest install base for AI-powered PCs and workstations. With the ability to handle complex language models, Windows now empowers developers and users to explore new frontiers in natural language processing and AI-powered applications.

Introducing NVIDIA Omniverse and Omniverse Cloud

5.1 Industrial Digitalization with Omniverse

NVIDIA Omniverse is a groundbreaking platform for industrial digitalization, offering unprecedented capabilities for creating, simulating, and visualizing complex environments. By leveraging virtualization and real-time rendering technologies, Omniverse enables industries to design and optimize their operations in a highly immersive and collaborative manner. From architecture and automotive to manufacturing and entertainment, Omniverse unlocks new possibilities for industries to digitally transform their processes, revolutionizing the way they operate and innovate.

5.2 Availability of Omniverse on Azure Cloud

Building on the partnership with Microsoft, NVIDIA has made Omniverse available on Azure Cloud. This expansion allows industries to harness the power of Omniverse's digitalization capabilities through the flexibility and scalability of Azure's cloud infrastructure. By hosting Omniverse on Azure, NVIDIA and Microsoft are enabling businesses of all sizes to access and leverage the transformative potential of industrial digitalization. This integration not only streamlines workflows but also accelerates the pace of innovation and collaboration, driving digital transformation across diverse industries.

AI Foundry Service: Enabling Enterprise AI

6.1 The Opportunity of Generative AI

Generative AI has unlocked new opportunities for enterprises to engage with artificial intelligence like Never before. Its usefulness, versatility, and ease of use have made AI more accessible to businesses worldwide. Companies can now leverage generative AI in multiple ways, driving innovation, enhancing operations, and delivering value to their customers. The collaboration between NVIDIA and Microsoft has played a pivotal role in democratizing AI, making it accessible to organizations across various sectors.

6.2 Three Ways Companies can Utilize Generative AI

The availability of generative AI through public cloud services, embedded applications, and custom-built models offers companies a multitude of options to utilize AI capabilities. Firstly, public cloud services, such as Chad GPT, enable organizations to leverage pre-built AI models and solutions, providing immediate access to powerful AI capabilities. Secondly, embedded AI applications, such as Windows, open up new possibilities for integrating AI directly into existing software and workflows. Finally, companies have the option to Create their own AI models and solutions tailored to their specific requirements, utilizing their own data and creating proprietary large language models. These three approaches provide companies with the flexibility to choose the most suitable path to engage with AI and drive innovation within their organizations.

NVIDIA's Three Pillars for Custom Large Language Models

7.1 Leveraging AI Expertise, AI Enterprise, and AI Factories

The development of custom large language models requires advanced AI expertise, comprehensive AI enterprise solutions, and state-of-the-art AI factories. NVIDIA offers these three pillars to help customers build and deploy their own proprietary language models. Leveraging NVIDIA's AI expertise and end-to-end workflows, companies can rely on the industry's best practices and guidelines for efficient model development. The AI Enterprise solution provides a comprehensive platform for deploying and managing AI models at scale, ensuring high performance and reliability. Additionally, NVIDIA's AI factories, available on Azure as djx Cloud, provide a powerful and scalable infrastructure for training cutting-edge language models, similar to how TSMC acts as a foundry for chip manufacturing.

7.2 Building Proprietary Language Models with NVIDIA

The introduction of an AI Foundry Service by NVIDIA allows companies to realize the full potential of custom large language models. By leveraging the three pillars offered by NVIDIA, organizations can embark on their AI Journey with confidence, knowing they have access to top-tier AI expertise, enterprise-grade solutions, and powerful AI factories. This level of support and infrastructure empowers companies to create their own proprietary language models tailored to their specific needs. NVIDIA's commitment to providing a comprehensive AI ecosystem, paired with the collaborative efforts of Microsoft, ensures that companies can unlock the full potential of AI and drive innovation in their respective industries.

The Arc of AI Innovation

8.1 Generative AI as a Platform Transition

Generative AI represents the most significant platform transition in the history of computing. Compared to previous advancements like personal computers, mobile devices, and the internet, generative AI is set to have an even greater impact. The scope of generative AI extends far beyond these innovations and stands as the largest TAM (total addressable market) expansion in the computer industry. With the emergence of dedicated AI data centers, the stage is set for a new era of computing solely focused on running AI models and generating intelligence. This paradigm shift marks the birth of AI factories and paves the way for unprecedented advancements in AI-driven technologies.

8.2 Significance of AI Factories and Co-Pilots

Two critical components shaping the future of AI innovation are AI factories and co-pilots. AI factories represent a new class of data centers dedicated to AI computation, enabling businesses and researchers to harness the true power of AI. These factories, combined with the software counterpart - co-pilots, provide a new level of accessibility and scalability for AI development. Co-pilots leverage existing tools and infrastructure to enhance the development process, acting as intelligent assistants to guide and accelerate AI model creation. The combination of AI factories and co-pilots creates an ecosystem ripe for innovation, powering the next Wave of AI advancements.

8.3 Three Waves of AI Innovation

The arc of AI innovation can be divided into three distinct waves. The first wave witnessed the rise of startups and cloud internet services, which developed breakthrough technologies and services in the field of AI. The Second wave, currently underway, is characterized by the integration of AI into enterprise-grade applications and services, exemplified by Microsoft's co-pilot initiatives and Windows 365. As businesses begin to embrace AI technologies on a larger scale, this wave marks a turning point in AI adoption. The third and largest wave is anticipated to revolve around heavy industries that heavily rely on AI for digitalization. NVIDIA's Omniverse, combined with generative AI, will drive this wave, enabling heavy industries to unlock the transformative power of AI and accelerate their digitalization efforts.

The Microsoft-NVIDIA Partnership

9.1 Transforming Industries through Collaboration

The partnership between Microsoft and NVIDIA is founded on a deep collaboration that has transformed industries and Shaped the technology ecosystem. This relationship goes beyond a traditional supplier-buyer dynamic, focusing on advancing fundamental computer science and driving innovation. The values of collaboration, partnership, and ecosystem building are deeply ingrained in the Microsoft-NVIDIA partnership, resulting in powerful advancements such as confidential computing and generative AI. Moreover, Microsoft's hosting of NVIDIA's native stack on Azure has amplified the reach and impact of NVIDIA's ecosystem, fostering a thriving community of developers and innovators worldwide.

9.2 Advancing Computer Science and Inventing New Technologies

The Microsoft-NVIDIA partnership is driven by a joint commitment to advancing computer science and inventing new technologies. The collaboration has led to remarkable achievements, from the development of the world's fastest AI supercomputer to the integration of AI capabilities into Windows and the introduction of NVIDIA Omniverse and Omniverse Cloud. By combining their expertise and resources, Microsoft and NVIDIA have pushed the boundaries of what is possible in AI and computing. With a customer-centric approach and a focus on innovation, the partnership continues to set new benchmarks and Shape the future of technology.

Highlights

• The collaboration between NVIDIA and Microsoft has led to significant progress and achievements in the field of AI and computing.
• The development of the world's fastest AI supercomputer demonstrates the power of accelerated computing and the speed at which technological advancements can be made.
• Microsoft's integration of AI capabilities into Windows has made it a leading platform for large language models.
• NVIDIA Omniverse and Omniverse Cloud enable industries to digitally transform and innovate in unprecedented ways.
• The AI Foundry Service offered by NVIDIA empowers companies to build their own custom large language models.
• The future of AI innovation lies in the advancements of AI factories and the utilization of co-pilot tools.
• The Microsoft-NVIDIA partnership is characterized by collaboration, innovation, and the advancement of computer science.
• The partnership has resulted in significant achievements such as confidential computing and generative AI.

FAQ

Q: What is accelerated computing? A: Accelerated computing is a full stack challenge that involves offloading general-purpose computing to accelerate software performance. It encompasses various aspects, including computing, networking, and the transformation of the computer industry through generative AI.

Q: How did the collaboration between NVIDIA and Microsoft result in the world's fastest AI supercomputer? A: The collaboration between NVIDIA and Microsoft allowed for the planning and implementation of the world's fastest AI supercomputer within a few months. This achievement showcases the power of accelerated computing and the capabilities of both teams.

Q: How does Windows become a first-class client for large language models? A: By integrating NVIDIA's AI technology, Windows has become a leading platform for large language models. This integration opens up new possibilities for developers and users, enabling them to leverage the power of AI seamlessly.

Q: What is NVIDIA Omniverse, and how does it benefit industries? A: NVIDIA Omniverse is a platform for industrial digitalization that offers capabilities for creating, simulating, and visualizing complex environments. It revolutionizes industries by providing immersive and collaborative tools for optimizing operations and driving innovation.

Q: How can organizations utilize generative AI? A: Companies can utilize generative AI in three main ways: through public cloud services like Chad GPT, embedding AI into applications like Windows, and building their own proprietary large language models using their data.

Q: What are NVIDIA's three pillars for custom large language models? A: NVIDIA offers the pillars of AI expertise, AI Enterprise solutions, and AI factories for building custom large language models. These pillars provide the necessary support and infrastructure for companies to develop their own language models.

Q: What is the significance of AI factories and co-pilots in AI innovation? A: AI factories and co-pilots play a crucial role in AI innovation. AI factories represent dedicated data centers for AI computation, while co-pilots act as intelligent assistants to guide and accelerate AI model development.

Q: How does the Microsoft-NVIDIA partnership transform industries? A: The partnership between Microsoft and NVIDIA has transformed industries through collaboration, innovation, and the development of breakthrough technologies. It has resulted in advancements such as confidential computing and generative AI.

Q: What is the vision for the future of the Microsoft-NVIDIA partnership? A: The Microsoft-NVIDIA partnership aims to Continue advancing computer science and inventing new technologies. Both companies are committed to driving innovation, fostering collaboration, and shaping the future of technology.