Intel Investor Conference 2022: Announcing technology roadmaps and key milestones

Source: Intel

At the 2022 Investor Conference, Intel shared its product and process technology roadmap and key milestones to drive growth across business units.

At Intel's 2022 Investor Conference, Intel CEO Pat Kissinger and heads of various business units outlined the company's development strategy and long-term growth plan. In an era of strong demand for semiconductors, Intel's long-term plans will fully grasp the opportunities for transformational growth. In the presentation, Intel unveiled its product roadmap and key execution nodes for its key business units, including:

Data Centers and Artificial Intelligence

Client-side computing

Accelerate computing systems and graphics

Intel Foundry Services

Software and advanced technology

Network and edge

Technological progress

Intel Data Center and Artificial Intelligence (DCAI) has announced a roadmap for next-generation Intel Xeon products to be released between 2022 and 2024. Intel will build a robust data center ecosystem with industry-leading hardware and software capabilities and lead new advances in artificial intelligence and security. Intel's data center program will not only enable it to gain a new share in fast-growing markets such as artificial intelligence, networking, and system encryption, but also increase data center revenue with industry-leading Xeon products.

Intel Xeon Product Roadmap Update

Intel's next-generation Xeon product roadmap includes:

Sapphire Rapids – Starting in the first quarter of 2022, Intel will deliver the Sapphire Rapids processor manufactured on intel's 7-process process and bring to market Intel's most feature-rich Xeon processor to date. Sapphire Rapids will dramatically improve performance across a wide range of workloads, delivering up to 30x performance gains in AI alone.

Emerald Rapids – Emerald Rapids, scheduled to hit the market in 2023, is a next-generation Xeon processor manufactured on intel's 7-process process that will enhance performance while further enhancing the memory and security benefits of existing platforms.

● New architectural strategy – Future generations of Xeon will have a dual-track product roadmap based on both the P-core and the energy efficiency core to integrate the two optimized platforms into a common, industry-defining platform. This new architecture strategy will maximize the performance per watt and segmentation capabilities of the product, thereby comprehensively enhancing Intel's overall competitiveness in the industry.

Sierra Forest – In 2024, Intel will introduce the revolutionary new energy-efficient Xeon processor Sierra Forest, a leading product based on Intel's 3-process process with high density and ultra-energy-efficient performance.

Granite Rapids – Based on confidence in intel's 3 process, Intel announced that it will upgrade the process of granite Rapids processors from Intel 4 to Intel 3. This next-generation Xeon performance core processor product will be available in 2024, further strengthening Intel's overall leadership in the industry.

Intel Client Computing Division (CCG) provides an overview of the client products to be introduced in the coming years. As PCs become more important than ever, Intel expects the Client Computing Division to continue to be a major driver of Intel's growth. In 2021, Intel's global PC shipments exceeded 340 million, up 27% from 2019. Intel expects the PC market demand to remain strong and continue to grow, mainly due to the growing demand for replacement of existing customer bases and the increase in PC ownership and penetration rates worldwide.

Client Computing Division Roadmap Updates

Building on today's dominant products, Intel's next-generation client product roadmap includes:

Raptor Lake – Raptor Lake, which is expected to go public in the second half of 2022, will bring a two-digit performance boost and enhanced overclocking capabilities compared to Alder Lake. The Raptor Lake configuration will be upgraded to up to 24 cores and 32 threads, using Intel 7 process and a high-performance hybrid architecture. Raptor Lake will be compatible with the sockets of the Alder Lake system.

Meteor Lake and Arrow Lake – Meteor Lake will use the Intel 4 process. Arrow Lake will be the first product to be built on Intel 20A wafers, as well as chips manufactured using an external manufacturing process. These two products, which combine artificial intelligence and graphics card chips that bring independent graphics card-level performance, will achieve a huge breakthrough in the direction of XPU. Meteor Lake will be available in 2023, and Arrow Lake will be officially launched in 2024.

Lunar Lake and more— To advance the IDM 2.0 strategy, Intel will develop best-in-class products using both internal and external process nodes.

Three sub-divisions of the Accelerated Computing Systems and Graphics Division (AXG) are shipping products as planned and are expected to generate more than $1 billion in revenue for the company in fiscal 2022. As Intel's growth engine, the three sub-divisions of the Accelerated Computing Systems and Graphics division are expected to generate nearly $10 billion in revenue by 2026.

Visual Computing Product Roadmap and Strategic Planning

● Intel Iris Graphics Timing Nodes and Roadmap Updates - Accelerated Computing Systems & Graphics is expected to ship more than 4 million discrete GPUs in 2022. OEMs will release laptops configured with Intel Iris graphics (codenamed Alchemy) in the first quarter of 2022. Intel will ship discrete graphics for desktops in the second quarter and discrete graphics for workstations in the third quarter. In addition, Celestial, for the super enthusiast market, has officially begun its architecture research and development work.

● Endgame Project - Through the Endgame Project service, users can directly read Intel Iris graphics card information and get a long-time access and low-latency computing experience. The Endgame project will officially go live later this year.

Supercomputing product roadmap and strategic planning

At present, more than 85% of the world's supercomputers use Intel Xeon processors. On this basis, the Accelerated Computing Systems & Graphics division will further enable higher computing power and memory bandwidth, and deliver an industry-leading CPU and GPU product roadmap to empower high-performance computing (HPC) and AI workloads. As of now, Intel expects to receive more than 35 HPC-AI designs from leading OEMs and cloud service providers (CSPs). In addition, the Accelerated Computing Systems and Graphics Division has developed a technical route to Z-level computing by 2027.

● Sapphire Rapids with built-in high-bandwidth memory (HBM) – Saphire Rapids with built-in high-bandwidth memory are capable of providing up to 4x the memory bandwidth for applications, achieving up to 2.8x performance improvements compared to 3rd generation Intel Xeon Scalable processors. In addition, in the same computational fluid dynamics application scenario, saphire Rapids with built-in high-bandwidth memory is 2.8 times better than similar solutions.

Ponte Vecchio – Accelerated Computing Systems and Graphics will deliver Ponte Vecchio GPUs for the Aurora supercomputer project later this year. Faced with complex financial services workloads, Ponte Vecchio meets industry-leading performance standards and demonstrates 2.6x performance over market-leading solutions.

Arctic Sound-M – Arctic Sound-M is Intel's first GPU equipped with a hardware AV1 encoder, which delivers a 30% increase in bandwidth based on it, as well as an industry-leading open source media solution. A supercomputer for media and analytics that will enable high-quality transcoding, stream density, and cloud gaming. Arctic Sound-M is now customer-facing and is expected to ship in mid-2022.

Falcon Shores – Falcon Shores is a new architecture that integrates x86 and Xe graphics cards in the same slot. The architecture, which is scheduled to be available in 2024, will deliver more than 5x performance improvements per watt in terms of performance per watt, compute density, memory capacity, and bandwidth.

Custom computing department

The Custom Computing Division, part of the Accelerated Computing Systems & Graphics Division, will develop custom products for a number of emerging workloads, including blockchain, edge supercomputing, high-end in-vehicle infotainment systems, and immersive displays.

As cars become more electric, safer, smarter and more connected than ever before, the automotive industry is currently undergoing a profound transformation. These trends are driving considerable growth, with revenue in the automotive semiconductor industry doubling and expected to reach $115 billion by 2030. Today, incomplete supply chains and traditional process technologies will not be able to support the growing demand and the transition to more compute-intensive applications. To that end, Intel Foundry Services (IFS) is assembling a dedicated automotive team to provide complete solutions for automakers, focusing on the following three areas:

Open central computing architecture

Intel Foundry Services (IFS) will develop a high-performance, open automotive computing platform that helps automotive OEMs build next-generation experiences and solutions. This open computing architecture will leverage chiplet-based building blocks, along with Intel's advanced packaging technology, to provide significant flexibility for building solutions optimized for technology nodes, algorithms, software, and applications to meet the computing needs of the next generation of autonomous vehicles.

Automotive-grade foundry platform

Intel will enable manufacturing technology to meet the stringent quality requirements of automotive applications and customers. Intel Foundry Services (IFS) aims to combine advanced process and technology optimization with advanced packaging for microcontrollers and unique automotive needs to help customers design multiple types of automotive semiconductors. Mobileye's extensive experience with automotive-grade products as a leader in advanced driver assistance systems (ADAS) solutions enables Intel Foundry Services (IFS) to deliver advanced process technologies to automotive customers.

Realize the transition to advanced technology

Intel Foundry Services (IFS) will provide design services and Intel's IP to automakers, enabling them to leverage Intel's expertise from chip to system design. The automotive projects of Intel's Foundry Services Accelerator Program, announced last year, are designed to help automotive chipmakers transition to advanced process and packaging technologies and innovate with Intel's custom, industry-standard-based IP portfolio.

Software is a key component of Intel's competitive advantage, adding value to the entire software stack of Intel's client, edge, cloud, and data center businesses. Intel's approach is to foster an open ecosystem that ensures trust, selectivity, and interoperability within the industry, and serves as a catalyst for technology adoption and innovation. Intel's investment in software also presents disruptive and transformative growth opportunities.

Cross-platform, open development

The Intel OneAPI Toolkit provides a cross-platform, open programming model that empowers developers to solve unique challenges with optimized performance.

Solve challenges with artificial intelligence

The convergence of security and artificial intelligence demonstrates great promise for open and collaborative frameworks that help protect data while gaining insights. Intel Core processors and Intel vPRO systems use Intel Threat Detection Technology (Intel TDT) to detect malware behavior under the operating system and provide these insights to endpoint detection and response solutions. For confidential computing in the cloud, 3rd generation Intel Xeon processors with Intel Software Guard Extensions (Intel SGX) protect data and AI models, so they aggregate data and gather deeper insights to solve challenging problems, such as identifying brain tumors.

Networking and edge computing are evolving rapidly. To accelerate growth and drive the shift to software-defined and fully programmable infrastructure, Intel established the Networking and Edge Division (NEX) in 2021. Intel expects the revenue growth of the networking and edge business to exceed the growth rate of the total potential market in the next decade, which will make an important contribution to the company's overall growth. To seize this opportunity, the Networking & Edge Business Unit is rolling out programmable hardware and open software from the cloud to the Internet and 5G networks to the intelligent edge.

Intelligent structure

The Intel Smart Fabric Programmable Platform enables customers to program end-to-end network behavior through infrastructure in the data center, driving business opportunities and putting control in the hands of customers, providing them with a way to program their networks. This allows customers to continuously evolve, improve, and differentiate their infrastructure according to their needs. In addition, it creates a future with a whole new type of computing device, the Infrastructure Processing Unit (IPU). The computing device can be integrated into the data center, accelerating cloud infrastructure development and maximizing performance.

Mobile Network Transformation

For more than a decade, Intel has been leading the transformation of telecom networks and driving global networks to break free from the shackles of traditional fixed-function hardware and enable them to be defined by open, interoperable software. Intel's ambition is to provide customers with the industry's cutting-edge, broad programmable platform to drive business opportunities and put control in the hands of developers to support the expansion of 5G and more advanced technologies.

Accelerate the intelligent edge

Intel helps customers deliver intelligent edge platforms by offering a diverse portfolio of hardware and software products and a vast partner ecosystem. Intel Networking & Edge supports new use cases and workloads across a range of vertical industry markets designed to meet the growing demand for compute and analytics at the intelligent edge. Artificial intelligence—especially reasoning at the edge—can deliver beneficial insights in place, in real time, where and where data is generated. As a result, it is becoming the most common use case in edge computing, transforming and automating places like factories, smart cities, hospitals, and more.

Intel expects to once again lead the industry in transistor per watt performance by 2025. Intel's advanced test and packaging technologies give us unique industry leadership that benefits our products and foundry customers and plays a key role in advancing Moore's Law. Continuous innovation is the cornerstone of Moore's Law, and innovation is everywhere at Intel.

Process

With the launch of the 12th generation Intel Core processors, as well as other products coming soon in 2022, the Intel 7 is in production and shipping in bulk. The Intel 4 will use extreme ultraviolet lithography (EUV) technology and is expected to go into production in the second half of 2022, with about 20% more transistor performance per watt. The Intel 3 will have more features and an approximately 18 percent improvement in performance per watt, with production expected in the second half of 2023. Kicking off the Amy era with two technologies, The HybridFET and PowerVia, the Intel 20A will achieve an approximately 15% increase in performance per watt and will be operational in the first half of 2024. The Intel 18A will achieve an approximately 10% improvement in performance per watt and is expected to come into production in the second half of 2024.

encapsulation

Intel's leadership in advanced packaging technology provides designers with options across thermal, power, high-speed signal, and interconnect density to maximize and optimize product performance. In 2022, Intel expects to deliver leading packaging technologies on Sapphire Rapids and Ponte Vecchio, with trial production on Meteor Lake. Foveros Omni and Coveros Direct are advanced packaging technologies announced by Intel at Intel's Accelerated Innovation: Process Process and Packaging Technologies Online Launch in July 2021 and are expected to come into production in 2023.

innovation

When looking ahead to technologies such as High-NA EUV, RibbonFET, PowerVia, Foveros Omni, and Coveros Direct, Intel realizes that innovation never ends, so Moore's Law will continue to move forward. Intel is expected to provide about a trillion transistors in a single device by the end of this century, and we are working tirelessly to achieve this goal.