As you heard from Pat Gelsinger and Sandra Rivera today, Intel is embarking on a bold new path that expands our data center leadership across all workloads, delivers ultra-efficient compute to our customers and will enable our industry’s data center future. Offering our customers the ability to harness leadership performance-per-watt and performance-per-core with one unified Xeon platform is something only Intel can deliver. It’s this unmatched combination of features and our underlying software ecosystem that powers the majority of the world’s data centers today.
Today we’re sharing that future generations of Intel® Xeon® will have a new dual-track roadmap of Performance-core and Efficient-core based products, moving from two optimized platforms into one common, industry-defining platform. This new path will maximize performance-per-watt, segment features and Intel’s overall competitiveness within the industry. The first generation of these products, code-named Sierra Forest, is an Efficient-core (E-core) Xeon CPU being developed in parallel with our existing Performance-core (P-core) Xeon, Granite Rapids.
But how did we arrive here?
Over the past two years, in close partnership with our customers, we have been hard at work examining our strategic assets from both hardware and software perspectives. We’ve been running countless tests and performance projections across different architectural options. This tireless effort has paid dividends as we will deliver leadership performance across key design points on the Xeon platform.
As part of this work, we are also seeing customers take a variety of approaches when it comes to how they are optimizing their infrastructure. Some are optimizing for workloads that benefit from high performance-per-core and low latency, like database, AI and HPC type workloads. Others are focused on maximizing performance-per-watt to support a growing number of dense, latency-tolerant workloads that are mostly task parallel in nature, like front-end web services and data analytics.
Knowing this, let’s take a deeper dive into why the addition of an E-core-based Xeon product line is such a big deal for Intel, our customers and the industry. E-cores have been specifically designed to deliver power--efficient and high-density throughput performance that supports cloud native deployments. With multiple E-cores fitting into the physical space occupied by one P-core, we are able to deliver higher density at the socket and rack level.
Although we are introducing a new core architecture into our Xeon Platform product line, we were intentional about ensuring it remained compatible with the vast Xeon platform ecosystem that exists today. But these next-generation processors will share more than just platform, software and BIOS. Within the CPU package, we will decouple core and uncore functions into “compute tiles” and “I/O tiles,” with the I/O tiles being common between P-core and E-core based products, enabling a common I/O subsystem to be used.
This reduces complexity for both Intel and our customers who are building and validating systems. Additionally, the interconnect used between cores and memory controllers is the same. The beauty of this is that by simply swapping a compute tile, our customers get an expanded portfolio of benefits from a single platform investment.
Sierra Forest and our future E-core Xeon roadmap is an exciting innovation for us and for the industry. Combined with the rest of our future Xeon roadmap, Intel’s product, platform and packaging technology advancements will unleash our customers’ ability to shape future cloud-based architectures.
This single platform will help shape the data centers of the future to be more workload- focused and sustainable — and it’s coming to you in 2024 on our own Intel 3 process.
Stay tuned for additional updates on this exciting new product line in the coming months.
Lisa Spelman is corporate vice president and general manager of the Xeon and Memory Group at Intel Corporation.