Intel Corporation (INTC) Earnings Call Transcript & Summary

Welcome, welcome, welcome. Welcome to the launch of Intel's 4th Generation Xeon processor and our MAX family. Today isn't just about announcing a product. It's about showing how customers industry are already taking advantage of this amazing technology because this launch is very real, right? OEMs, our cloud providers, our developers are already embracing the 4th Gen Xeon and MAX family of processors. And the 4th Gen and the MAX family delivers extraordinary performance gains, efficiency, security capabilities, breakthrough our new capacities in AI and cloud and networking and delivering the world's most powerful supercomputers that have ever been built. And it truly is a pleasure to be back in person and interacting with you and participating in today's announcement, but it's also great to hear from our industry colleagues. They're people that we worked with for decades. We've been partners in technology here with us. And we want to hear about how they're excited about today's announcement and today's technology. And after we finish up with that, then we're going to hear from our very own, Sandra Rivera, as she launches the 4th Gen and MAX series of processors. So with that, let's hear from our industry partners now. Thank you so much for being here in person for today's announcement and for those online today is a great day. Play the video. [Presentation]

Please welcome the Executive Vice President of the Data Center in AI group, Sandra Rivera.

Hello, and welcome to the launch of our latest generation of data center solutions. Today, it represents an important day for Intel and for our customers. And I'm delighted that you're joining us to celebrate this moment with us. For over 2 decades, Intel technology has represented innovation in data centers and networking equipment and the cloud computing infrastructure across the globe. We worked closely with many of you to solve the greatest computing challenges at scale. And today, we're building upon that foundation of innovation by bringing to market a highly differentiated portfolio of data center solutions that will enable us to create and innovate the next generation of solutions. So businesses continue to evolve and grow. You are the critical partners working with us to invent and deliver the future of technology. When I joined Intel in 2000, the cloud was giving us an early glimpse of how technology would impact all of our lives. Today, data is being used to transform entire industries. In health care, medical researchers are using thousands of data set from patients in order to identify and predict brain tumors. In agriculture, farmers are using autonomous in order to boost crop yield. And in retail, online search data is being used to serve up better recommendations for consumers. The world, as we know it, is evolving before our eyes, and it's exciting to see our customers tackling these complex challenges and pushing the boundaries of innovation. As organizations look to scale, drive down costs and generate new revenue streams, it's more important than ever that businesses have the technology they need that drives the highest business value. For 3 generations of Intel's scalable processors, we've architected our data center solutions with your business needs in mind. Our unique approach is focused on providing real-world, purpose-built workload acceleration that delivers superior system performance at greater efficiencies. And purpose-built workload acceleration is not just about adding more cores. It requires a true system-level approach in which highly optimized software is tuned ti the differentiated features in our hardware in order to accelerate the most critical business workloads, including AI, networking, HPC and security. And today, we're taking purpose-built workload acceleration to the next level. I'm excited to unveil Intel's 4th Gen Xeon scalable processor. She's a beauty. These processors access represent a paradigm shift in the way businesses run their workloads and solve their computing challenges. 4th Gen Xeon features the most accelerators built into a data center processor in the industry, providing leading performance in workloads that matter most. We've demonstrated an average of 2.9x increase in performance for [indiscernible] efficiency across the group of the most common workloads. And these aren't just industry benchmarks, these are real-world workload performance schemes. And when we look at total cost of operations, with acceleration, customers can expect to see TCO improvements ranging from 52% to 66%. These step function performance gains are only possible because of the differentiated features we built into 4th Gen Xeon. These allow customers to make optimal use of core resources for greater CPU utilization and improved power efficiency. For workloads like AI, customers can achieve up to 10x higher pipe torch, real-time inference and training performance using Intel advanced matrix extensions. And the networking applications, customers can encrypt data using Intel quickest technology to achieve up to 47% fewer cores at the same performance level. This represents the future of purpose-built workload acceleration. And we could only achieve these gains with close collaboration with the world's leading software providers, OEMs and ODMs. There are thousands of software engineers at Intel working with our ecosystem partners to optimize and validate applications to run best on Intel hardware, and Xeon-based infrastructure is available virtually everywhere in the world. Today, there are over 100 million Xeons installed in the market from on-prem data centers to 5G networking equipment to edge infrastructure and, of course, in cloud service provider services. And within the cloud, Xeon is the most ubiquitous processor in the world. Xeon-based services are available virtually every cloud service provider in every global region. And this pervasiveness ensures customers get consistent performance and reliability when running their workloads in the cloud. Let's hear from AWS about the new business services they're offering based on our latest data center solutions.

I'm Dave Brown, Vice President of Amazon EC2. It's a pleasure to be here for the launch of the 4th generation Intel Xeon scalable processes. AWS Intel have been on a journey together innovating for customers since AWS introduced the concept of cloud computing with Amazon EC2 in 2006, over 16 years ago. Since then, customers have launched over 30 billion EC2 instances. AWS has built the most reliable and secure global cloud infrastructure with the broadest and deepest portfolio of instances. Amazon EC2 provides customers with the most options for compute, so they can choose the right infrastructure for their application and business needs. To date, AWS has introduced over 400 Intel-based instances, optimized for a wide variety of applications. At our recent Reinvent Conference, we launched our latest Intel-based instances, including network optimized instances as well as an optimized offering for HPC workloads. Customers such as NASDAQ, [ FanDuel Group ], Dana Farber Cancer Institute and Nissan Motors are examples of customers who have benefited from the performance, reliability, availability, security and scale of AWS infrastructure with Intel Xeon scalable processors for their applications. AWS is excited to continue our collaboration with Intel with our support for 4th Generation Intel Xeon scalable processors for the Amazon EC2 portfolio. Amazon EC2 R7iz instances already available today use the 4th Generation Intel Xeon scalable processes and have the highest performance per VCPU among our x86-based EC2 instances. They also deliver up 20% higher performance than comparable high-frequency instances and are ideal for electronic design automation, relational databases, data analytics and other workloads requiring a combination of high compute performance and high memory. Intel is also a customer of AWS, including running their electronic design automation workloads on AWS. And we are excited for Intel to utilize the latest Intel Xeon scalable processors on AWS for these workloads. Congratulations to the Intel team on this important launch milestone. We at AWS look forward to our ongoing collaboration with Intel to deliver more value for AWS customers.

With 400 Intel-based instances, Xeon is the most widely deployed platform in AWS' infrastructure, allowing customers to access the unique features in our hardware from every region in the world. And in addition to AWS, many of the world's leading cloud service providers already have 4th Gen Xeon services in public preview and general availability. These include Alibaba, Baidu, ByteDance, Google Cloud, IBM Cloud, Microsoft Azure, Oracle Cloud and Tencent. I think you'll be interested in what they have to say. [Presentation]

When I talk with our cloud partners, data security is top of mind. And as a continuance landscape evolves to become more distributed with workflows moving across ecosystems and multi-talent environments, protecting data at every set of workflow is paramount, whether it's data at rest, in transit or in use. Intel and ecosystem partners pioneered a data privacy technology called confidential computing. This is to enable businesses to handle their sensitive data and cloud environments. Intel is the only silicon provider to offer application level isolation with Intel software guard extensions, which provide the smallest attack surface for confidential computing, whether that's in private public or edge computing cloud environments. And now with 4th Gen Xeon, we're delivering level of virtual machine isolation technology called Intel trust domain extensions. This technology is ideal reporting existing applications to run in a confidential computing cloud environment. Microsoft Azure, Alibaba, Google Cloud and IBM Cloud will be the first to offer Intel TDX ahead of general availability. Here to share more about how they're using this innovative technology is Microsoft. [Presentation]

Thank you, Mark. Intel and Microsoft have been great partners in delivering impactful technology to the world for many years. And speaking of impactful technologies, artificial intelligence is most disruptive and transformative technology and computing today. AI is helping industries across the globe, process data, make sense of and make better decisions from massive amounts of data being generated. For generations, we have built-in AI accelerators into the world's most widely deployed data center processor. And with workloads, we are working with the ecosystem to ensure that AI frameworks and AI software is optimized for Intel hardware and is readily accessible. This work is helping us democratize AI and easy to deployment across all industries. My team has played a major role in Intel's AI journey and to talk more about how we're making this transformative technology broadly available, please welcome Corporate Vice pre and General Manager of Intel on Products, Lisa Spelman.

It's good to be here.

It's great to have you here, Lisa. So how many young launches have you now been a part of?

Okay. Well, unbelievably, this is my eighth Xeon launch, but I have to say that every single one gets more exciting. You mentioned that democratization of AI. And that's possible because of the accessibility of Xeon. This has been multiyear journey for us, working with our customers and our partners. In fact, in 2017, that was when we first introduced built-in AI acceleration to Xeon. And since then, we haven't stopped. Every generation, we're adding to that hardware acceleration, but more importantly, even we are working really hard and closely together with the software ecosystem to deliver that out-of-box ease and performance. So it goes up a notch with the 4th Gen. We're taking it to the next level. Everything that's been added to 4th Gen Xeon is because of our customers' input and their requirements. That ubiquity of Xeon plus the artificial in performance that customers now can access means that we're delivering today this wonderful opportunity for them to deliver both deep learning training and inference on the CPU that they already have in their environment.

With this progress, we've reached a tipping point with widespread deployment of artificial intelligence on Intel architecture. We recently set down with Fujitsu to talk about our work together and how AI is accelerating their digital transformation.

Let's listen in. [Presentation]

So [ Uro ], thank you so much for joining us here today.

Yes. Thank you, and it's really an honor to be here.

So Uro, what were your technology requirements defining the platform that you're offering to customers?

On one hand, of course, low latency because when customers are using the platform, it has to be very responsive. Energy consumption, because this is playing a major role. We are facing climate change everywhere. Energy consumption has to be reduced. And I think, therefore, the new Gen 4 Xeon platform will be a great step forward in this respect. And what we have seen with the Gen 4 Xeon is really now the capability that you have dedicated accelerators in place. So we have accelerators for AI, which is very important for us. So making sure that we can bring the workflow to dedicated unit on the chip and really making sure that the CPU can do whatever they want and having support from the framework perspective, so which means support from TensorFlow, PyTorch, OpenVINO, [ AMX net ]. We see also with the new 4th Generation of the Xeon processors and enormous speed up because with the tile technology and that you have dedicated accelerators to multiply those metrics, making sure that the data can be moved very quickly from one point to the other. And this allows us really to squeeze the most out of the CPU. And what we see in most of the cases, the CPU is more than good enough to do training and of course inference as well. And this is what we are targeting for.

Yes. Well, that was part of the architecture decisions that we made, which is how can we incorporate more acceleration into the base CPU platform. So what are the ways customers are using the platform?

One of the projects I really would like to highlight because I'm involved in this project is ASFINAG. ASFINAG is the company's operating and maintaining highways in Austria. What we would like to do with this R&D project we are doing with ASFINAG is really detecting electric vehicles. And unfortunately, there is no standardization in the European Union. In Germany, we have a E at the end of the number plate. In Austria, it's a green color of the characters of the number plate, so which means we have to detect an attribute of the character, which is the green color, but there are also some emblems on the plate itself. And this is not an easy task, but we have seen in the first stage that we were capable to do 2.3 detections per second. This was okay-ish. And then we have a chat with the colleagues here from Intel and they said, "Yes, we'll improve this," and then they have done an improvement with OpenVINO. And all of a sudden, we are now capable to achieve a detection rate of 30 images per second with the existing Xeon technology and with Gen 4, we are expecting a significant improvement because the numbers that we have seen from Intel, they're really unbelievable.

All great examples of where we are having a positive impact in driving that time to value, accelerating that pace of innovation. What are you excited about in terms of the future?

What we see right now is there is a hurdle from a customer point of view for the first step into AI. And I would call the new Xeon generation really democratizing AI because you must have a server in place already. And when you have all of those capabilities, you have accelerators, it's super fast and you can handle the data, you can use cryptography, compression, decompression and all of those aspects.

Yes. Well, our purpose, our mission, to democratize AI, to drive outcomes that improve the lives of every person on earth. Uro, thank you again for with us today and for sharing so many great examples of the high-impact work we're driving together and how we use technology for good.

Yes. And thank you for the opportunity of being here.

This is just one example of our customers already building solutions on fourth gen Xeon. And they're taking advantage of that acceleration, AMX, the crypto acceleration and also that software that I referenced. With this improved performance, you can deliver deep learning and inference workloads like natural language processing, recommendation systems, image recognition. Gen-over-gen, we delivered 55% lower total cost of ownership for real-time AI inference workloads. It's impressive performance paired with TCO that makes Xeon that industry foundation for inference. And with AMX, Xeon is quite capable for training as well. Xeon has been the foundation for scalable high-performance computing for decades. For the first time, Intel is integrating high bandwidth memory into the Xeon processor and delivering Xeon MAX. This innovative product addresses the #1 constraint in high-performance computing solutions. And it dramatically increases the memory bandwidth that's available and accelerate key high-performance computing workloads, like life and material sciences, manufacturing, high-energy physics. For years, compute capacity has grown at a rate much faster than the memory bandwidth. And this has led to a situation where you have workload performance that doesn't keep up. It's stranded compute. We haven't been feeding the cores enough data. Well, this has been an obstacle to progress. It leads to wasted compute cycles, wasted energy and costs. Xeon MAX is the first and only x86-based processor with integrated high-bandwidth memory. And for all your software developers out there, one of the most exciting things is that you can take advantage of all of that bandwidth without having to do code changes. The product brings 3.7x performance improvement across real-world workloads and requires 68% less energy than deployed competitive systems. Xeon Max provides you with the opportunity to maximize your bandwidth, maximize your compute and maximize developer productivity. Now Jim Lujan from Los Alamos National Labs is going to share with us a bit about how they're solving real-world problems using both Xeon fourth gen and Xeon MAX.

Jim, it's great to see you. So can you tell me a little bit about how the lab uses high-performance computing to advance your mission?

Absolutely. It's good to be here, [ Jeff ]. Thank you for having me. So at Los Alamos, we use modeling and simulation for a variety of areas, primarily stockpile stewardship, but we also do modeling and simulation and epidemiology, planetary defense, climate modeling.

Now what are some of the key bottlenecks that you experienced in those workloads?

So for us, we've done an analysis on some of our key applications. Our applications, while they do a lot of compute, are not as restricted by compute as they are with memory. So when we put together the crossroads procurement, we were looking for ways to improve time to insight to really reduce those bottlenecks on memory bandwidth. And Intel's Xeon MAX processor with that kind of high-bandwidth memory was really exciting for us because it really does improve that time to insight that we were looking for.