When in Rome: AMD Announces New Epyc CPU for HPC, Server and Cloud Wins

By John Russell

September 18, 2019

Where else but Rome could AMD hold the official Europe launch party for its second generation of Epyc microprocessors, codenamed Rome. Today, AMD did just that announcing key server wins, important cloud provider wins – including IBM Cloud – introduction of a new 64-core, 280-watt CPU aimed at high-end HPC, and indulging in some chest-thumping around performance metrics. And that’s not all. Yesterday, Dell EMC announced five new servers based on second-gen Epyc.

You get the picture. For AMD recently, it’s been (almost) all sunshine and haymaking. There’s also been a noticeable shift in positioning. When AMD plunged back into the datacenter market in 2017, contrition for past error and emphasis on competitive TCO (cum solid technology) were the main talking points. That’s changing as AMD seeks to add technology leadership to the message mix and cash in on its delivery of 7nm promises and product roadmap while Intel stumbles a bit.

In his media pre-briefing yesterday, Scott Aylor tackled technology achievement first. “One of the major themes in San Francisco” at the official launch of the Rome line roughly six weeks ago, “was [our] 80 World Records and counting,” he said. “We’re excited to say that number is going to be eclipsed. We don’t have the final count but know it will be north of 100.”

Last week at Tabor Communication’s conference, HPC &AI on Wall St., Aylor told HPCwire, “We think about pivoting from a TCO oriented play in the market to a leadership performance position that, by the way, has a fantastic TCO.” Rome, AMD hopes, is the platform to start accomplishing that goal.

Here’s a tech snapshot: The second-gen Epyc SoCs are built as nine-die packages with eight 7nm complex core die (CCD) chiplets — with up to eight cores each — surrounding a 14nm I/O die, connected via AMD’s second-gen Infinity fabric. AMD says the next-generation Infinity architecture gives customers access to the most I/O and memory bandwidth in its class, with 128 lanes of PCIe generation 4 (or more in custom builds). The upgraded fabric nearly doubles throughput, pushing 18GT/s socket-to-socket compared with 10.7 GT/s of throughput in the first generation chips.

It’s still good to remember AMD has a smaller waterfront to guard than larger rival Intel and that AMD’s market share remains in single digits although the company has aspirations for double digits in a year or so. In terms of technology, there wasn’t much new revealed at today’s Europe event that hadn’t been disclosed at the US launch. (For a deeper dive into second-gen Epyc’s technology, see HPCwire coverage, AMD Launches Epyc Rome, First 7nm CPU).

For the HPC community, release of the new sku – 7H12 – is perhaps most interesting news. It is nearly the same as the 7742 part but has 280W TDP (+55W); 2.6 GHz base (+350 MHz); 3.3 GHz boost (-100 MHz). The new sku is designed to leverage water cooled systems and is already used by Atos in the BullSequana XH 2000. Aylor noted it’s a “hugely dense platform with the ability to put eight processors essentially in 1U form factor.” The BullSequana XH2000 supports other microprocessors as well, including Intel and Arm (Marvell ThunderX2), and Nvidia GPUs for accelerators.

Atos BullSequana XH2000 supercomputer

Atos has lined up customers for Epyc-based versions of the BullSequana XH2000 – CSC, Finland (7-plus petaflops system), Uninett, Norway (6 petaflops), and GENCi, France (12 petaflops). Moreover, Atos today announced publishing the top results for two-socket nodes on four SPECrate benchmarks and set a new record for the HPL Linpack Benchmark.

Atos issued the following statement: “We’re extremely proud that our BullSequana has achieved these world-record results. Our unique Enhanced Direct Liquid Cooling system provided the most efficient environment for achieving such performance of the AMD EPYC processor,” said Agnès Boudot, SVP, head of HPC & Quantum at Atos. “Our BullSequana equipped with the latest AMD chip, provides our customers with the highest available performance for HPC and AI workloads, with an optimized TCO, to support them in going beyond the limits of traditional simulation.”

Supermicro will also offer systems based on the 7H12 (Ultra & GPU platforms) which are scheduled to be available in Q4 2019. The 7H12 will be broadly available to the market according to Aylor.

On the cloud front, AMD announced IBM Cloud would offer Epyc-based instances. “We have been in in close partnership with IBM Cloud for a long time,” said Aylor. In terms of matching second-gen Epyc to IBM needs, he said, “Think about things like very large container implementations, very high performance bare metal implementations, as well as virtualized environments.”

OVHcloud, a dominant player in Europe, also announced it would use second-gen Epyc. “[They will offer] the 7402P as part of their dedicated hosting environment, leveraging a full flash implementation with direct connect via NVMe to the 7402P which is a single socket solution. And again, looking at driving this in terms of virtualization, so containers really optimized on greatest level of density and thinking about computing economics,” said Aylor. Availability of these instances is expected by EOY 2019.

AMD’s first-gen Epyc has already enjoyed wide success among cloud providers, particularly in pushing 1-socket solutions. The hope is the second-gen devices will do as well. Google and Azure have announced plans to use second-gen Epyc. In his pre-briefing Aylor set high expectation for further adoptions, “You can also see some of the folks that have been on stage with us and can kind of assume where they’re headed. It’s also worth noting that we have a China event in the middle of October where I think there will be substantial news based upon some of the major players in China.”

The Dell EMC news – released yesterday and also discussed at today’s AMD event – had been expected. It included two single-socket and three dual-socket PowerEdge servers based on second-gen Epyc. HPCwire’s sister pub, EnterpriseAI reported yesterday: “With an eye on HPC going mainstream, Ravi Pendekanti (Dell EMC SVP, server infrastructure) said the high-end, two-socket model achieved 3,462 Gigaflops, representing a more than 200-percent performance upgrade based on the Linpack benchmark. Those performance benchmarks grew out of earlier development at Dell EMC’s recently formed HPC and AI Innovation Lab.” (For more see coverage, Dell’s AMD-Powered Server Line Targets High-End Jobs)

The last prominent win previewed by Aylor was Nokia’s decision to use second-gen Epyc processors. AMD reported Nokia will use Epyc to “accelerate its Cloud Packet Core system which helps service providers deliver converged broadband, IoT, and machine-type communication services for 5G.” According to AMD, during testing Nokia was able to get a 2X increase in packet throughput compared to previous systems.

Given the seemingly rapid market traction for second-gen Epyc, there have been reports of concern around meeting demand. AMD seemed to be addressing that by announcing a multi-generation partnership with its fab TSMC (Taiwan Semiconductor Manufacturing Co.) as well as announcing TSMC’s decision to use second-gen Epyc in its own datacenter.

In his pre-briefing Aylor said, “[The multi-generational partnership that we have around seven nanometer [is] a key catalyst for us having this leadership capability in terms of cache density, in terms of core density, and a very compelling level performance. You will hear from TSMC that they have a full commitment and full capability to ramp the volumes that are needed in the market. They believe in the technology so much that they will actually be deploying it in their datacenter. So, TSMC will also talk about how they are now deploying Epyc second generation Rome in their data center for virtualization environments for EDA deployments.”

Let the good time roll for now.

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!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen 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…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. 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…

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…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a 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…

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…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

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…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t 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…

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…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire