AZ-800 learn - Administering Windows Server Hybrid Core Infrastructure Updated: 2023

November 16, 2023, 10:10 AM EST

‘Microsoft copilot is that one experience that runs across all our surfaces, understanding your context on the web, on your device and when you’re at work, bringing the right skills to you when you need them,’ says Microsoft CEO Satya Nadella.

Microsoft Chairman and CEO Satya Nadella declared an “age of copilots” and introduced a host of new technology and products aimed at making the vendor No. 1 in artificial intelligence from the infrastructure layer to data and applications.

Copilots—Microsoft’s name for its generative AI tools that can generate text, code and other content—are the new user interface “that helps us gain access to the world’s knowledge and your organization’s knowledge,” Nadella saidduring his keynote address at Redmond, Wash.-based Microsoft’s annual Ignite conference.

“Most importantly, it’s your agent that helps you act on that knowledge,” Nadella said.

[RELATED: Microsoft Ignite 2023: The Biggest News In AI, Copilot]

Nadella Talks AI Infrastructure

On the subject of infrastructure, Nadella called Microsoft “the world’s best systems company across heterogeneous infrastructure” with a data center network that is “one of the most advanced and extensive in the world already.”

Microsoft’s hollow core fiber (HCF) technology—part of the reason for the vendor’s acquisition of Lumenisity in December—helps Microsoft deliver a 47 percent improvement in network speed, Nadella said.

Microsoft’s position as a systems company led to one of the vendor’s biggest announcements at Ignite 2023: its first custom, in-house CPU series, Azure Cobalt. Cobalt already powers parts of Microsoft Teams, Azure Communication Services and Azure SQL. The chips will become available to users next year, Nadella said.

When talking about Azure’s capabilities, Nadella called out its MLPerf benchmarking consortium submission of 10,000 H100 GPUs—three times more than the previous record.

Top500 ranked the Eagle system installed in Azure as the most powerful supercomputer in the public cloud and third including private clouds. Ahead of Eagle was the No. 1 Frontier system based on Hewlett Packard Enterprise Cray architecture and equipped with AMD Epyc processors. The No. 2 Aurora system is based on HPE Cray with Intel Xeon processors.

“That made news,” Nadella said. “What didn’t make news is we didn’t submit the entirety of our supercomputer. We just submitted only a fraction of our supercomputer. So I’m thrilled to be No. 3 with that.”

Nadella also showcased the Microsoft Azure Maia AI accelerator, complete with an end-to-end rack to help with AI compute workloads’ cooling and networking density needs.

“This architecture allows us to take this rack and put it into existing data center infrastructure and facilities rather than building new ones,” he said.

Nadella On AI Models, Data

A major unveiling around the models that power GenAI was Nadella describing a models-as-a-service offer in Azure. Models-as-a-service will allow “access to these large models that are all available as open source as just hosted APIs without you as developers having to provision GPUs so that you can focus on development, not back-end operations,” he said.

He promised Azure AI users the latest in developments from Microsoft-backed OpenAI, the organization behind the ChatGPT text generator and Dall-E image generator.

“As OpenAI innovates, we will deliver all of that innovation as part of Azure AI,” he said. “And we are bringing the very latest of GPT-4—including GPT-4 Turbo, GPT-4 Turbo with Vision—to Azure OpenAI Service. … GPT-4 Turbo will be available in Azure open AI Service this week in preview. And the token pricing for the new models will be at parity with OpenAI.”

In data, Microsoft is “integrating the power of AI across the entirety of the data stack,” Nadella said.

He called the launch of the Microsoft Fabric data platform—now generally available—“our biggest data launch perhaps since SQL Server.” Fabric already has 25,000 customer users.

Fabric’s mirroring capability “is a frictionless way to add existing cloud data warehouses as well as databases to Fabric from Cosmos DB or Azure SQL DB—as well as Mongo and Snowflake—not only on our cloud, but any cloud to Fabric,” he said.

Microsoft added vector indices to Cosmos DB and PostgreSQL, he said. The vendor also moved the management of AI-powered indices out of the application domain and into the database itself with Azure AI extensions for PostgreSQL.

“This makes it easy and efficient for developers to use AI to unlock the full potential of all their relational data in the database,” he said.

He called Microsoft’s Azure AI Search “a first-class vector search plus state-of-the-art reranking technology, delivering these very high-quality responses much beyond what you can just get from a vanilla vector search.”

Microsoft The Copilot Company

Nadella said that Microsoft is “the copilot company” and described “a future where there will be a copilot for everyone and everything you do.”

“Microsoft copilot is that one experience that runs across all our surfaces, understanding your context on the web, on your device and when you’re at work, bringing the right skills to you when you need them,” he said. “Just like, say, today you boot up an operating system to access applications or a browser to navigate to a website, you can invoke a copilot to do all these activities and more—to shop, to code, to analyze, to learn, to create. We want the copilot to be everywhere you are.”

Users can leverage plugins and GPTs—“a new way for anyone to create a tailored version of ChatGPT that’s more helpful for very specific tasks at work or at home,” Nadella said—to customize copilots with the new Copilot Studio.

“It can connect to databases, custom back ends, legacy systems that may even be on-premises,” he said. “All of this allows you to extend copilot with capabilities unique to your organization and the systems you use every day.”

AI With Mixed Reality, Quantum Computing

The Microsoft CEO ended his keynote with a taste of the next stage in AI: AI with mixed reality and AI with quantum.

Microsoft’s Copilot in Dynamics 365 (D365) Guides, unveiled during Ignite 2023, shows how “the real world becomes your prompt and interface,” Nadella said.

“Pay attention to how not just your voice, but your gestures—even where you look—becomes a new input and how transformative it can be to someone like a frontline worker using Dynamics 365,” he said.

The technology isn’t science fiction, with Siemens and Chevron among the Microsoft customers previewing it. “It’s great to see the power,” he said. “And I think it’s going to be even more powerful in the years to come.”

For AI and quantum computing users, Nadella said that Azure Quantum Elements can generate new chemical compounds for simulations just as GenAI applications can generate text.

“Just imagine if you can compress 250 years of progress in chemistry and material science into the next 25 years,” he said. “That’s truly using the power of AI to change the pace of science. In this example, I’m just using a Python notebook—I mean, think about it, just a Python notebook—with Quantum Elements to discover a new coolant, a process that would have taken three years if they just used traditional computational techniques, but it probably takes nine hours now.”

He continued: “I can reason over these results with a copilot, narrow them down, find the most promising candidates. Using Quantum Elements, any scientist can design novel new molecules with unique properties for developing more sustainable chemicals, drugs, advanced materials or batteries. And this is just the very beginning.”

Nadella highlighted Microsoft’s partnership with Photonic to bring quantum computing to networking.

“Combining that infrastructure and bringing it right into Azure takes us one more step closer to the promise of quantum networking and computing inside of Azure,” Nadella said. “At the end of the day, though, all of this innovation will only be useful if it’s empowering all of us in our careers, in our communities, in our countries. … Just imagine if 8 billion people always had access to a personalized tutor, a doctor that provided them medical guidance, a mentor that gave advice for anything they need. I believe all of that’s within reach.”

As it ramps up its artificial intelligence (AI) strategy, Microsoft has developed a custom semiconductor, the Maia AI Accelerator, which it will use internally for its Copilot and Azure OpenAI service. There are also plans to make the technology available to Azure customers.

Among the main drivers for developing Maia is greater efficiency and optimisation when servers based on the new semiconductor are used in Microsoft Cloud datacentres.

According to Brian Harry, a Microsoft technical fellow leading the Azure Maia team, the Maia 100 AI Accelerator has been designed specifically for the Azure hardware stack. He claimed this vertical integration – which aligns chip design with the larger AI infrastructure designed for Microsoft’s workloads – can yield huge gains in performance and efficiency.

Sam Altman, CEO of OpenAI, said: “Since first partnering with Microsoft, we’ve collaborated to co-design Azure’s AI infrastructure at every layer for our models and unprecedented training needs. We were excited when Microsoft first shared their designs for the Maia chip, and we’ve worked together to refine and test it with our models. Azure’s end-to-end AI architecture, now optimised down to the silicon with Maia, paves the way for training more capable models and making those models cheaper for our customers.”

Along with Maia AI, the company also unveiled the Microsoft Azure Cobalt CPU, an Arm-based processor tailored to run general purpose compute workloads on the Microsoft Cloud along with expanding industry partnerships to provide more infrastructure options for customers. These partnerships include a new NC H100 v5 Virtual Machine Series built for Nvidia H100 Tensor Core GPUs and the addition of Nvidia H200 Tensor Core GPU to its fleet next year to support larger model inferencing with no increase in latency.

Microsoft has also added AMD-powered AI hardware to its portfolio in ND MI300 virtual machines, which it said are designed to accelerate the processing of AI workloads for high-range AI model training and generative inferencing. ND MI300 uses AMD’s latest GPU, the AMD Instinct MI300X.

Scott Guthrie, executive vice-president of Microsoft’s Cloud + AI Group, said: “Microsoft is building the infrastructure to support AI innovation, and we are reimagining every aspect of our datacentres to meet the needs of our customers. At the scale we operate, it’s important for us to optimise and integrate every layer of the infrastructure stack to maximise performance, diversify our supply chain and deliver customers infrastructure choice.”

There is growing evidence that the computational power required to run AI workloads and, in particular, large language models, consumes huge amounts of energy.

In February, research firm SemiAnalysis suggested OpenAI required 3,617 of Nvidia’s HGX A100 servers, with a total of 28,936 graphics processing units, to support ChatGPT, implying an energy demand of 564 MWh per day. The research estimated that ChatGPT costs $694,444 per day to run.

A latest New York Times article quoted a researcher who predicts the electricity needed to run AI could increase the world’s carbon emissions significantly, but this would depend on whether the datacentres are powered by fossil fuels or renewable energy resources.

Wes McCullough, corporate vice-president of hardware product development, said choosing Arm technology was a key element in Microsoft’s sustainability goal.

“The architecture and implementation is designed with power efficiency in mind,” he said. “We’re making the most efficient use of the transistors on the silicon. Multiply those efficiency gains in servers across all our datacentres, it adds up to a pretty big number.”