Intel Plans Falcon Shores 2 GPU Supercomputing Chip for 2026  

By Agam Shah

August 8, 2023

Intel is planning to onboard a new version of the Falcon Shores chip in 2026, which is code-named Falcon Shores 2. The new product was announced by CEO Pat Gelsinger during last week’s earnings call. 

“We’re already working on Falcon Shores 2 for 2026,” Gelsinger said. 

The Falcon Shores 2 chip will succeed the first Falcon Shores chip — which is a high-performance GPU for AI and supercomputing — in 2025.  

“We have a simplified road map as we bring together our GPU and our accelerators into a single offering,” Gelsinger said. 

Gelsinger was also referring to the AI chips code-named Gaudi. Gaudi2 AI is already being sold to customers and will be followed by the Gaudi3 chip. 

Gelsinger did not share further details about Falcon Shores 2, but the company may reveal more information about it at next month’s Intel Innovation developer show. 

Intel’s enterprise GPU roadmap has taken multiple hits in the last six months as the company implements austerity measures to become a manufacturing-first company. 

It started toward the end of 2022 with the dismantling of the original Accelerated Computing Systems and Graphics Group (AXG) group, which was formed in 2021 and was led by Raja Koduri, who left the company in March. 

AXG’s consumer graphics portion was folded into the client computing group, and enterprise accelerated computing moved over to the data center group. 

In March, Intel chopped many data center chips and GPUs. Intel also scrapped plans for its original 2024 Falcon Shores chip, which was planned as an “XPU”, or an integrated CPU and GPU. Falcon Shores was then repositioned as a discrete GPU to be released in 2025. 

Intel also axed a GPU codenamed Rialto Bridge, which was due next year as a successor to the current Max Series GPU codenamed Ponte Vecchio. The GPU was scheduled to go into Barcelona Supercomputing Centre’s MareNostrum 5 supercomputer. 

Intel also cut the upcoming data center GPU called Lancaster Sound, which was the successor to last year’s Flex Series GPU codenamed Arctic Sound. The chipmaker is moving to a two-year upgrade cycle for the Flex GPUs, and the successor is now Melville Sound. 

Intel has not talked much about its 2026 product lineup, and Falcon Shores 2 is among the first chips to be revealed for that timeframe. Intel changed its plans for the original Falcon Shore as the market response for XPUs was cold, and there was healthy demand for discrete GPUs. 

The Falcon Shores chips will use chiplets, so Intel will be able to mix and match GPUs, AI accelerators, and third-party CPUs. 

Falcon Shores2
Falcon Shores 2

“This provides the flexibility across vendors to marry Falcon Shores GPU with other CPUs as well as the CPU to GPU ratio,” said Jeff McVeigh, corporate vice president and general manager of the Super Compute Group at Intel, during a call in May. 

The discrete-GPU Falcon Shores product model uses a common GPU-based programming interface, and the CXL interface from the CPU and GPU will improve productivity and performance of code, McVeigh said.  

The release of Falcon Shores 2 just a year after the first version could be Intel’s response to keep up with its rivals. Intel is currently behind AMD, which will start shipping the MI300 accelerator, and Nvidia offers the red-hot H100 GPUs.  

The 2026 release date implies Falcon Shores 2 will also be made using the Angstrom-era process. The chip maker has focused on product releases until 2025, when it will achieve its goal of starting five nodes in four years. 

The original Falcon Shore XPU was to be made using the Angstrom manufacturing process, which could be Intel’s latest 18A or 20A process, which will begin in 2024 and 2025.  

But the 2026 release date for Falcon Shores 2 gives a short lifecycle to the original Falcon Shores GPU, which could be a bridge product. Gelsinger has not been shy in cutting short server product lifecycles to get to the latest and greatest product. 

Rialto Bridge was originally designed to be a bridge product to the Falcon Shores XPU before it was axed. Intel’s upcoming Emerald Rapids Xeon CPU is an incremental upgrade to the current Sapphire Rapids chip, with next year’s Granite Rapids being a major next-generation upgrade. 

The Falcon Shores chips are designed for HPC and AI computing, and Intel has talked about merging the GPU with the Gaudi chip lineup. 

The demand for GPUs has shot up with the emergence of generative AI. Tesla’s Elon Musk talked about a shortage of Nvidia GPUs during the company’s most recent earnings call. Companies are looking at alternatives in AI chips from Intel, AMD, and Cerebras, which scored a multimillion-dollar deal to ship its AI chips to G42, a Middle Eastern cloud provider. 

Gelsinger said during the earnings call said Falcon Shores is being “well executed,” and that Gaudi2 and Gaudi3 were poised to do well.  

“The accelerator pipeline is now well over $1 billion and growing rapidly, about 6x this past quarter. That is led by, but not exclusively, Gaudi, but also includes the Max and Flex product lines as well,” Gelsinger said. 

Intel’s first quarter earnings, which was reported in April, had a surprise – it recorded supplementary revenue from data center GPUs, said Dean McCarron, principal analyst at Mercury Research. 

The revenue was not previously reported, and likely from the sales of Ponte Vecchio, McCarron said. 

But Intel’s immediate AI opportunity is in Gaudi2, which is shipping, and Gaudi3, which is coming soon. Intel’s goal is to make sure the AI software stack is forward compatible on Gaudi and Falcon Shores chips via its OneAPI software stack. 

“We’ll just be broadening the flexibility of that software stack. We are adding FP8. We just added PyTorch 2 support. Every step along the way, it gets better and broader use cases. More language models are being supported. More programmability is being supported in the software stack,” Gelsinger said. 

Falcon Shores has the “best of GPU and the best of matrix acceleration,” Gelsinger said 

“But every step along the way, it just gets better. Every software release gets better. Every hardware release gets better along the way to cover more of the overall accelerator marketplace,” Gelsinger said. 

Intel is taking a partially open software approach with OneAPI, which is centered around a tool called SYCLomatic, a software tool that can convert proprietary CUDA code to run on a wide range of GPUs, including Ponte Vecchio.   

Intel’s server roadmap is looking healthy with products being released ahead or on schedule. The PC and server chip launches are aligned with the start of the new processes.  

Emerald Rapids will be released on Intel 7 this year, a PC chip called Meteor Lake on Intel 4 next year, server chips Sierra Forest and Granite Rapids on Intel 3 next year, a PC chip called Arrow Lake will lead off Intel’s 20A next year, and Clearwater Forest will lead off the most advanced 18A node in 2025. Intel’s PC chip called Lunar Lake will likely be made using 18A. 

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

ISC 2024 Takeaways: Love for Top500, Extending HPC Systems, and Media Bashing

May 23, 2024

The ISC High Performance show is typically about time-to-science, but breakout sessions also focused on Europe's tech sovereignty, server infrastructure, storage, throughput, and new computing technologies. This round Read more…

HPC Pioneer Gordon Bell Passed Away

May 22, 2024

Legendary computer scientist Gordon Bell passed away last Friday at his home in Coronado, CA. He was 89. The New York Times has a nice tribute piece. A long-time pioneer with Digital Equipment Corp, he pushed hard for de Read more…

ISC 2024 — A Few Quantum Gems and Slides from a Packed QC Agenda

May 22, 2024

If you were looking for quantum computing content, ISC 2024 was a good place to be last week — there were around 20 quantum computing related sessions. QC even earned a slide in Kathy Yelick’s opening keynote — Bey Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Core42 Is Building Its 172 Million-core AI Supercomputer in Texas

May 20, 2024

UAE-based Core42 is building an AI supercomputer with 172 million cores which will become operational later this year. The system, Condor Galaxy 3, was announced earlier this year and will have 192 nodes with Cerebras Read more…

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's latest weapon in the AI battle with GPU maker Nvidia and clou Read more…

ISC 2024 Takeaways: Love for Top500, Extending HPC Systems, and Media Bashing

May 23, 2024

The ISC High Performance show is typically about time-to-science, but breakout sessions also focused on Europe's tech sovereignty, server infrastructure, storag Read more…

ISC 2024 — A Few Quantum Gems and Slides from a Packed QC Agenda

May 22, 2024

If you were looking for quantum computing content, ISC 2024 was a good place to be last week — there were around 20 quantum computing related sessions. QC eve Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's l Read more…

Europe’s Race towards Quantum-HPC Integration and Quantum Advantage

May 16, 2024

What an interesting panel, Quantum Advantage — Where are We and What is Needed? While the panelists looked slightly weary — their’s was, after all, one of Read more…

The Future of AI in Science

May 15, 2024

AI is one of the most transformative and valuable scientific tools ever developed. By harnessing vast amounts of data and computational power, AI systems can un Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

ISC 2024 Keynote: High-precision Computing Will Be a Foundation for AI Models

May 15, 2024

Some scientific computing applications cannot sacrifice accuracy and will always require high-precision computing. Therefore, conventional high-performance c Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

Leading Solution Providers

Contributors

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

The GenAI Datacenter Squeeze Is Here

February 1, 2024

The immediate effect of the GenAI GPU Squeeze was to reduce availability, either direct purchase or cloud access, increase cost, and push demand through the roof. A secondary issue has been developing over the last several years. Even though your organization secured several racks... Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh Read more…

Intel Plans Falcon Shores 2 GPU Supercomputing Chip for 2026  

August 8, 2023

Intel is planning to onboard a new version of the Falcon Shores chip in 2026, which is code-named Falcon Shores 2. The new product was announced by CEO Pat Gel Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

How the Chip Industry is Helping a Battery Company

May 8, 2024

Chip companies, once seen as engineering pure plays, are now at the center of geopolitical intrigue. Chip manufacturing firms, especially TSMC and Intel, have b Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire