Advanced Micro Devices (AMD) has once again demonstrated its commitment to pushing the boundaries of high-performance computing and artificial intelligence. The company's recent "Advancing AI 2024" event unveiled a slew of new products and partnerships that underscore AMD's ambition to become a leader in the rapidly expanding AI market.
The event, held in San Francisco, brought together industry leaders, partners, and developers to showcase AMD's latest innovations in data center technology, AI accelerators, and enterprise computing solutions. Dr. Lisa Su, AMD's Chair, President, and CEO, set the tone for the day by emphasizing the transformative power of AI across various sectors.
"Over the next decade, AI will enable so many new experiences that will make computing an even more essential part of our lives," Su stated. "AI can help save lives by accelerating medical discoveries. It can revolutionize research. It can create smarter and more efficient cities. It can enable much more resilient supply chains. And it can really enhance productivity across virtually every single industry."
Su outlined AMD's strategy for AI leadership, which revolves around four key pillars:
- Delivering high-performance, energy-efficient compute engines for AI training and inference
- Creating an open, proven, and developer-friendly software platform
- Fostering co-innovation with partners across the ecosystem
- Providing comprehensive solutions from chip to data center level
The AI Accelerator Market: A $500 Billion Opportunity
One of the most striking revelations from the event was AMD's updated forecast for the data center AI accelerator market. Su announced a significant upward revision of their previous estimates:
"Now as we look out over the next 4 years, we now expect the data center AI accelerator TAM will grow at more than 60% annually to $500 billion in 2028. For AMD, this represents just a huge growth opportunity."
This represents a substantial increase from their previous estimate of $400 billion by 2027, reflecting the explosive growth and demand in the AI sector.
5th Generation EPYC Processors: Turin Takes Center Stage
A major announcement at the event was the launch of AMD's 5th Generation EPYC processor family, codenamed "Turin." These new processors are built on the Zen 5 core architecture, which delivers impressive performance improvements:
"Turin is fantastic. It features up to 150 billion transistors across 17 chiplets, scales up to 192 cores and 384 threads. And one of the things that's very special about 5th Gen EPYC is we actually thought about it from the architectural standpoint in terms of how do we build the industry's broadest portfolio of CPUs that both covers all of the new cloud workloads as well as all of the important enterprise workloads."
Su presented two versions of Turin:
- A 128-core variant optimized for scale-up workloads with 16 4-nanometer chiplets
- A 192-core version optimized for scale-out workloads with 12 3-nanometer compute chiplets
Both use a 6-nanometer I/O die in the center.
The performance gains of the new EPYC processors are substantial:
"When you look at the competition's top of stack, a dual socket 4th Gen EPYC server is already 1.7x faster on SPECrate 2017. And with 5th Gen EPYC, the performance is fantastic. We extend that lead with 2.7x more performance."
In enterprise workloads, 5th Gen EPYC provides significant advantages:
"When running these workloads on-prem, again, 4th Gen EPYC is already the performance leader. And with 5th Gen, we deliver 1.6x more performance per core than the competition. That's 60% more performance with no additional licensing cost."
The business case for upgrading to 5th Gen EPYC is compelling:
"5th Gen EPYC can do that same amount of work of 1,000 servers in just 131 Turin servers. Now just think about what that means. It's like a huge benefit for CIOs. You can replace 7 legacy servers with 1 single EPYC server, and that significantly reduces the power that you need in your data center, it lowers TCO by more than 60%."
Instinct MI300 Series: Accelerating AI Workloads
AMD's Instinct MI300 family of AI accelerators, launched in December 2023, has seen significant adoption and performance improvements. Su reported:
"In the last 10 months, we've been laser-focused on ensuring our customers get up and running as fast as possible with maximum performance right out of the box. And to do that, we've had to significantly drive improvements and continuous improvements across our ROCm software stack."
The company introduced the MI325X, the next-generation Instinct accelerator:
"MI325 again leads the industry with 256 gigabytes of ultrafast HBM3E memory and 6 terabytes per second of bandwidth. When you look at MI325, we offer 1.8x more memory, 1.3x more memory bandwidth, 1.3x more AI performance in both FP16 and FP8 compared to the competition."
Su also previewed the future of AMD's AI accelerator roadmap:
"Today, I'm very excited to give you a preview of our next-generation MI350 series. The MI350 series introduces our new CDNA 4 architecture. It features up to 288 gigabytes of HBM3E memory and adds support for new FP4 and FP6 data types."
Partnerships and Ecosystem Development
Throughout the event, AMD emphasized the importance of partnerships and ecosystem development in driving AI innovation. The company showcased collaborations with major cloud providers, technology companies, and AI startups.
Amin Vahdat from Google Cloud highlighted their use of AMD EPYC CPUs:
"We've been using EPYC CPUs for multiple generations to serve our Cloud customers and our internal properties at Google. That adoption has been driven in large part by the gen-over-gen performance and efficiency gains we've delivered together."
Karan Batta from Oracle Cloud Infrastructure discussed their extensive deployment of AMD technologies:
"AMD and Oracle have been working together for a very long time. Since the inception of OCI in 2016, AMD EPYC CPUs are deployed across 162 data centers across the globe that covers our public cloud regions, our gov regions, our secret regions, even our dedicated regions in alloy as well."
Naveen Rao from Databricks shared insights on their collaboration with AMD:
"The large memory capacity and incredible compute capabilities of MI300X have been key to achieve over 50% increase in performance on some of our critical workloads."
Microsoft's partnership with AMD was highlighted through a conversation between Lisa Su and Satya Nadella, Microsoft's Chairman and CEO. Nadella stressed:
"We are very, very excited about your road map. Because at the end of the day, if I sort of get back to the core, what we have to deliver is performance per dollar per watt, because I think that's the constraint, right, which is if you really want to create abundance, right, where the cost per million tokens keeps coming down so that people can really go use what is essentially a commodity input to create higher value output, right?"
Kevin Salvadori from Meta revealed the extent of their AMD deployment:
"I'm happy to announce to everybody something you already know, that we've deployed over 1.5 million EPYC CPUs in Meta's global services."
Software and Developer Ecosystem
Vamsi Boppana, AMD's SVP of AI, detailed the progress made with ROCm, AMD's AI software stack:
"Over the last year, we've expanded functionality in ROCm at all layers of the stack at the lower layers, from coverage, for platforms, operating systems, to higher layers of the stack where we have expanded support for new frameworks like JAKs. We've implemented powerful new features, algorithms and optimizations to deliver the best performance for generative AI workloads."
Boppana announced a significant milestone in ecosystem support:
"Today, I'm delighted to say that over 1 million Hugging Face models now run on AMD. This has been made possible by our close collaboration over the last year, an effort that ensures that all their model architectures are validated on a nightly basis."
Enterprise PC Innovation
While the event focused heavily on data center and AI infrastructure, AMD also showcased innovations in enterprise PCs. The company introduced the Ryzen AI Pro 300 series processors:
"The AMD Ryzen AI Pro 300 series resets the bar for what a business PC can do. We combine our high-performance AM5 CPU, our new RDNA 3.5 graphics our new XDNA 2 NPU with 50-plus TOPS of AI performance, and all of this is within Copilot+ PCs."
Pavan Davuluri from Microsoft demonstrated new AI experiences enabled by these processors:
"Copilot+ PCs are the fastest, most intelligent, most secure Windows PCs we've ever built. And quite frankly, we're just energized with the feedback we received from customers and from reviewers. And we're excited to bring Copilot+ to Strix PRO power designs."
Conclusion
As Lisa Su reflected on her tenth anniversary as AMD's CEO, she emphasized the company's commitment to continuing innovation in high-performance computing and AI:
"You can count on all of us at AMD to continue to push the envelope on high-performance computing and AI, and we are just getting started."
The announcements and partnerships revealed at this event signal AMD's strong ambition to capture a significant share of the burgeoning AI market. As the industry continues to evolve at a rapid pace, AMD's strategy of open collaboration and continuous innovation positions it well to meet the challenges and opportunities that lie ahead in the world of AI and high-performance computing.
