By Jason Kimrey
Last week was an exciting one for Intel and our partners. At Intel Architecture Day, we built on our bold vision for a new generation of innovation and technology leadership.
If you didn’t take part in Intel Architecture Day 2021 last week — or missed the announcements — we detailed the biggest architectural shift in a generation for x86 central processing units (CPUs). These advances will pave the way for our next era of leadership products and these products will drive new business opportunities for the entire technology ecosystem.
At the event, Intel chief architect Raja Koduri shared the progress we’re making on our six pillars of technology innovation: software, security, interconnect, memory, XPU architecture, and process and packaging. This included a first in-depth view of Alder Lake, as well as exciting updates on Sapphire Rapids and Intel oneAPI.
So what does all this mean to you?
Path to Innovation
Architecture Day further demonstrates that Intel is making good progress on its innovation journey. We’re executing faster than ever before, which includes a renewed commitment to delivering leadership products across our portfolio for partners to enable their unique business needs.
Our execution engine is firing on all cylinders to deliver the reliable, secure and scalable products our partners need and expect from us. This means that over the next few years, our customers can count on Intel to deliver game-changing technologies and products needed to satisfy the crushing demand for more compute performance.
Looking ahead, these breakthrough announcements will deliver new capabilities in our client and data center products, as well as key initiatives to support interoperability across all platforms.
In the client product category, Intel continues to make game-changing advances to ensure we’re building the best-performing processors on the planet.
With Alder Lake, we announced the first performance hybrid architecture, which integrates a Performance-core and an Efficient-core to provide significant performance across all workload types. This will provide partners with the world’s most scalable system-on-chip (SoC) architecture, Intel Thread Director Technology, and pioneering next generation memory with Gen 5.
When Alder Lake becomes available by year’s end, it will deliver performance that scales to support all client segments, from ultra-portable laptops to enthusiast and commercial desktops.
We also announced a new discrete graphics microarchitecture, Xe-HPG, for gamers and creators. It’s been designed for scalability and enthusiast-class performance with a software-first approach. Products based on this microarchitecture will come to market in the first quarter of 2022 under the Intel Arc brand and Alchemist family of SoCs.
Intel is also driving towards unrivaled performance leadership in the data center. Sapphire Rapids, the next generation of Intel Xeon Scalable processors, represents the industry’s biggest data-center platform advancement in over a decade.
The Sapphire Rapids processor delivers substantial compute performance across dynamic and increasingly demanding data-center usages. And it’s workload-optimized to deliver high performance on elastic compute models like cloud, microservices and artificial intelligence (AI).
We also shared more insight on the Infrastructure Processing Unit (IPU), which is designed to enable cloud and communication service providers to reduce overhead and free up performance for CPUs. This includes multiple new IPU architectures to address the complexity of diverse and dispersed data centers.
Last but not least, we provided additional detail on delivering a world-class graphics processing unit (GPU) for the data center. Code-named Ponte Vecchio, it’s based on the Xe-HPC microarchitecture and will deliver industry-leading FLOPs and compute density to accelerate AI, high performance computing (HPC) and advanced analytics workloads.
Intel disclosed details of the Xe-HPC microarchitecture. Among them: Xe-HPC’s initial silicon performance is providing greater than 45 TFLOPS FP32 throughput, greater than 5 TB/sec. memory fabric bandwidth, and greater than 2 TB/sec. connectivity bandwidth.
To enable our customers to fully unlock the performance capabilities delivered by these world-class products, Intel last week provided additional insight on Intel oneAPI. This is a universal programming language that makes it easier to unleash application performance across multiple CPUs, GPUs, FPGAs and other accelerators.
We shared several incredible demos of how OneAPI enable next-generation performance for partners today. If you need more information on how to access to these products, or are looking to incorporate them into your software planning, please view the Intel oneAPI mini-site.
To sum it up, Intel’s best days are in front of us. We have excellent products coming to market, and as we continue executing against our roadmap, you can expect to see and feel the momentum of Intel’s innovation and product leadership.
You can trust that we are laser-focused on realizing our mission to deliver the next generation of products you rely on. And you can expect us to drive business opportunities within the ecosystem.
At Intel, we also know that we can’t do it alone. We appreciate your confidence and trust in Intel as your innovation partner.
Personally, I am excited about the work happening here at Intel. I hope these new developments are another reason for you and our other global partners to look forward to the many ways you can rely on Intel to enable your business.
Jason Kimrey is general manager of U.S. channel and partner programs at Intel.