AMD Charges Back into the Datacenter and HPC Workflows with EPYC Processor

By John Russell

June 20, 2017

AMD is charging back into the enterprise datacenter and select HPC workflows with its new EPYC 7000 processor line, code-named Naples, announced today at a “global” launch event in Austin TX. In many ways it was a full frontal assault on Intel’s dominance in the x86 datacenter landscape. Claiming performance and cost advantagaes and supported by statements from key OEMs, ODMs, and hyperscalers – HPE, Dell, and Microsoft Azure for example – AMD is hoping to convince HPC and datacenter customers it is back for the long haul.

Aware there may be market reluctance after its absence from the datacenter, Scott Aylor, AMD corporate VP and GM of enterprise solutions business, said “It’s not enough to come back with one product, you’ve got to come back with a product cadence that moves as the market moves. So not only are we coming back with EPYC, we’re also [discussing follow-on products] so when customers move with us today on EPYC they know they have a safe home and a migration path with Rome.” AMD has committed to socket compatibility between EPYC 7000 line and Rome, code name of the next scheduled generation AMD processor aimed at the datacenter.

AMD showcased some gaudy performance and price-performance benchmarks comparing EPYC to Broadwell line. In a pre-launch briefing with HPCwire, Aylor said, “These numbers are very big, so they show very measurable separation from what is available with Broadwell. Part of that is quite frankly because we didn’t design EPYC to compete with Broadwell. We designed it to compete with what’s coming. When [Intel’s] Skylake comes later this summer, we think these comparisons will still be very strong against the platinum, gold silver and bronze of Skylake.”

Based on the Zen core, EPYC is a line of system on a chip (SoC) devices designed with enhanced memory bandwidth and fast interconnect in mind. AMD also introduced a one-socket device, optimized for many workloads, which AMD says will invigorate a viable one-socket server market. With EPYC, “we can build a no compromise one-socket offering that will allow us to cover up to 50 percent of the two-socket market that is today held by the [Intel Broadwell] E5-2650 and below,” said Aylor.

AMD clearly has big ambitions. Earlier this spring it introduced Ryzen7 processor line, also based on the Zen core, and targeting high performance gaming. EPYC is aimed squarely at the datacenter. Aylor briefed HPCwire on EPYC before the launch and some of the technical details were still not available. It is an SoC product stack with a range of offerings roughly mimicking the Broadwell product stack. EPYC has up to 32 cores and 8 DDR4 channels per CPU allowing it to address 2TB of memory. The I/O is 128 PCIe lanes.

“The SoC approach we have taken allows all of the IO that has historically lived on an external bridge or IO hub to be fully integrated that into the device,” said Aylor. One result is low latency high performance connections. The PCIe lanes are configurable, “so you can use them to connect to SATA links, directly connect to NVMe links. It also facilitates a strong connection to high performance GPUs.” AMD plans to show an EPYC plus Radeon Instinct GPU machine learning platform at its conference this week.”

AMD presented both SPECint (integer) and SPECfp (floating point) performance comparisons with the Broadwell as well as price point comparisons (how much performance the same number of dollars will be of each processor) some of which are shown below.

“We’re tiering products in 32, 24, and 16-core ranges,” said Aylor. The idea, of course, is satisfy widely varying needs. The top end aimed at scale out and HPC workloads, he said. The bottom tier allows users to closely manage per core licensing costs. “We have tried to cover the vast majority of the market that exists today in the Broadwell family,” says Aylor. Every product will have a dedicated security processor.

“Sometime people will say benchmarks are interesting but how do you do in the real world. Well we will showcase a fluid dynamics HPC workload, Apache/Spark, and software defined storage reference architecture [at the launch]. We will also have an open stack cloud based implementation,” said Aylor. AMD was expecting on the order 600 attendees for the EPYC launch.

Moving back into the datacenter is a huge bet by AMD that’s required a very substantial investment in the Zen core and EPYC. Seeking to buttress the gamble, AMD has seemingly got buy-in from several market makers and many ecosystem partners. Here are four endorsements included in the official release; while the statements are on the over enthusiastic side they nonetheless suggest AMD has done productive groundwork with partners:

  • HPE. “The EPYC processor represents a paradigm shift in computing and will usher in a new era for the IT ecosystem,” said Antonio Neri, EVP and general manager Enterprise Group, HPE. “Starting with the Cloudline CL3150 and expanding into other product lines later this year, the arrival of EPYC in HPE systems will be welcomed by customers who are eager to deploy the performance and innovation EPYC delivers.”
  • Dell EMC. “As an industry leader, we are committed to driving IT Transformation for our customers,” said Ashley Gorakhpurwalla, president, server solutions division at Dell EMC, “Our next generation of PowerEdge servers are the bedrock of the modern datacenter that are designed to maximize business scalability and intelligent automation with integrated security. The combination of PowerEdge and the AMD EPYC performance and security capabilities will create unique compute solutions for our customers to accelerate workloads and protect their business.”
  • Baidu. “As the world’s largest Chinese language search engine and leading AI-Tech company, Baidu prides itself on simplifying a complex world through technology,” said By Dr. Zhang Ya Qin, president of Baidu. “The AMD EPYC processor powered one-socket server can significantly increase our datacenter computing efficiency, reduce TCO and lower energy consumption. We will start deploying with the launch of AMD EPYC and I look forward to our cooperation leading to scaled EPYC adoption this year, and ongoing innovations.”
  • Microsoft. “We’ve worked to make Microsoft Azure a powerful enterprise grade cloud platform, that helps guide the success of our customers, no matter their size or geography,” said Girish Bablani, corporate vice president, Azure Compute, Microsoft Corp. “To power Azure, we require the most cutting-edge infrastructure and the latest advances in silicon which is why we intend to be the first global cloud provider to deliver AMD EPYC, and its combination of high performance and value, to customers.”

The single socket gambit is another interesting aspect to AMD’s initiative. Currently two socket servers rule the roost.

Here’s the AMD pitch: “In our one socket offering we have come up with a clever way to maintain all of the I/O capabilities that you would get in a two socket as well as the full complement of eight memory channels. Today people buy two socket, sometimes because they need to, but more often than not because they have to. There are many examples in which I/O rich [workloads] like storage, like GPU compute, and some vertical workloads where people don’t necessarily need two sockets from a CPU performance perspective,” said Aylor.

AMD’s single socket offering cuts costs substantially, according to Aylor. “We’ve selectively optimized a couple of skews for one socket only. So these are skews that are one socket capable only,” said Aylor. As an example of how the one socket and two socket offerings are distinguished, he cited on package interconnect, “The infinity fabric that would normally connect the two sockets in a two socket system, we repurpose that interconnect into more I/O lanes and that’s how you have in a two socket solution 128 lanes of PCIe and in a one socket solution you still keep the same level of connectivity.”

AMD has singled out a number of vertical as good fits for one socket EPYC servers. Perhaps not surprisingly, storage is one. “Not only base line storage but software defined storage with EPYC’s ability to attach a massive number of SATA drives to a one socket. We also see a strong opportunity in certain areas of high performance computing, especially those that tend to focus on memory bound application. And we have an oil and gas reservoir simulation demo,” said Aylor.

Link to AMD press release: https://www.hpcwire.com/off-the-wire/amd-unveils-record-setting-epyc-datacenter-processor/

Subscribe to HPCwire's Weekly Update!

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

IBM Research Scales to 11,400 Cores for EDA

August 5, 2021

For many HPC users, their needs are not evenly distributed throughout a year: some might need few – if any – resources for months, then they might need a very large system for a week. For those kinds of users, large Read more…

Careers in Cybersecurity Featured at PEARC21

August 5, 2021

The PEARC21 (Practice & Experience in Advanced Research Computing) Student Program featured a Cybersecurity Careers Panel. Five experts shared lessons learned from more than 100 years of combined experience. While it Read more…

HPC Career Notes: August 2021 Edition

August 4, 2021

In this monthly feature, we’ll keep you up-to-date on the latest career developments for individuals in the high-performance computing community. Whether it’s a promotion, new company hire, or even an accolade, we’ Read more…

The Promise (and Necessity) of Runtime Systems like Charm++ in Exascale Power Management

August 4, 2021

Big heterogeneous computer systems, especially forthcoming exascale computers, are power hungry and difficult to program effectively. This is, of course, not an unrecognized problem. In a recent blog, Charmworks’ CEO S Read more…

Digging into the Atos-Nimbix Deal: Big US HPC and Global Cloud Aspirations. Look out HPE?

August 2, 2021

Behind Atos’s deal announced last week to acquire HPC-cloud specialist Nimbix are ramped-up plans to penetrate the U.S. HPC market and global expansion of its HPC cloud capabilities. Nimbix will become “an Atos HPC c Read more…

AWS Solution Channel

Pushing pixels, not data with NICE DCV

NICE DCV, our high-performance, low-latency remote-display protocol, was originally created for scientists and engineers who ran large workloads on far-away supercomputers, but needed to visualize data without moving it. Read more…

Berkeley Lab Makes Strides in Autonomous Discovery to Tackle the Data Deluge

August 2, 2021

Data production is outpacing the human capacity to process said data. Whether a giant radio telescope, a new particle accelerator or lidar data from autonomous cars, the sheer scale of the data generated is increasingly Read more…

Careers in Cybersecurity Featured at PEARC21

August 5, 2021

The PEARC21 (Practice & Experience in Advanced Research Computing) Student Program featured a Cybersecurity Careers Panel. Five experts shared lessons learn Read more…

Digging into the Atos-Nimbix Deal: Big US HPC and Global Cloud Aspirations. Look out HPE?

August 2, 2021

Behind Atos’s deal announced last week to acquire HPC-cloud specialist Nimbix are ramped-up plans to penetrate the U.S. HPC market and global expansion of its Read more…

What’s After Exascale? The Internet of Workflows Says HPE’s Nicolas Dubé

July 29, 2021

With the race to exascale computing in its final leg, it’s natural to wonder what the Post Exascale Era will look like. Nicolas Dubé, VP and chief technologist for HPE’s HPC business unit, agrees and shared his vision at Supercomputing Frontiers Europe 2021 held last week. The next big thing, he told the virtual audience at SFE21, is something that will connect HPC and (broadly) all of IT – into what Dubé calls The Internet of Workflows. Read more…

How UK Scientists Developed Transformative, HPC-Powered Coronavirus Sequencing System

July 29, 2021

In November 2020, the COVID-19 Genomics UK Consortium (COG-UK) won the HPCwire Readers’ Choice Award for Best HPC Collaboration for its CLIMB-COVID sequencing project. Launched in March 2020, CLIMB-COVID has now resulted in the sequencing of over 675,000 coronavirus genomes – an increasingly critical task as variants like Delta threaten the tenuous prospect of a return to normalcy in much of the world. Read more…

IBM and University of Tokyo Roll Out Quantum System One in Japan

July 27, 2021

IBM and the University of Tokyo today unveiled an IBM Quantum System One as part of the IBM-Japan quantum program announced in 2019. The system is the second IB Read more…

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make it seem like it's two nodes behind? For Intel, the response was to change how it refers to its nodes with the aim of better reflecting its positioning within the leadership semiconductor manufacturing space. Intel revealed its new node nomenclature, and... Read more…

Will Approximation Drive Post-Moore’s Law HPC Gains?

July 26, 2021

“Hardware-based improvements are going to get more and more difficult,” said Neil Thompson, an innovation scholar at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). “I think that’s something that this crowd will probably, actually, be already familiar with.” Thompson, speaking... Read more…

With New Owner and New Roadmap, an Independent Omni-Path Is Staging a Comeback

July 23, 2021

Put on a shelf by Intel in 2019, Omni-Path faced a uncertain future, but under new custodian Cornelis Networks, OmniPath is looking to make a comeback as an independent high-performance interconnect solution. A "significant refresh" – called Omni-Path Express – is coming later this year according to the company. Cornelis Networks formed last September as a spinout of Intel's Omni-Path division. Read more…

AMD Chipmaker TSMC to Use AMD Chips for Chipmaking

May 8, 2021

TSMC has tapped AMD to support its major manufacturing and R&D workloads. AMD will provide its Epyc Rome 7702P CPUs – with 64 cores operating at a base cl Read more…

Berkeley Lab Debuts Perlmutter, World’s Fastest AI Supercomputer

May 27, 2021

A ribbon-cutting ceremony held virtually at Berkeley Lab's National Energy Research Scientific Computing Center (NERSC) today marked the official launch of Perlmutter – aka NERSC-9 – the GPU-accelerated supercomputer built by HPE in partnership with Nvidia and AMD. Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer called Dojo to process truly vast amounts of video data. It’s a beast! … A truly useful exaflop at de facto FP32.” Read more…

Google Launches TPU v4 AI Chips

May 20, 2021

Google CEO Sundar Pichai spoke for only one minute and 42 seconds about the company’s latest TPU v4 Tensor Processing Units during his keynote at the Google I Read more…

CentOS Replacement Rocky Linux Is Now in GA and Under Independent Control

June 21, 2021

The Rocky Enterprise Software Foundation (RESF) is announcing the general availability of Rocky Linux, release 8.4, designed as a drop-in replacement for the soon-to-be discontinued CentOS. The GA release is launching six-and-a-half months after Red Hat deprecated its support for the widely popular, free CentOS server operating system. The Rocky Linux development effort... Read more…

Intel Launches 10nm ‘Ice Lake’ Datacenter CPU with Up to 40 Cores

April 6, 2021

The wait is over. Today Intel officially launched its 10nm datacenter CPU, the third-generation Intel Xeon Scalable processor, codenamed Ice Lake. With up to 40 Read more…

Iran Gains HPC Capabilities with Launch of ‘Simorgh’ Supercomputer

May 18, 2021

Iran is said to be developing domestic supercomputing technology to advance the processing of scientific, economic, political and military data, and to strengthen the nation’s position in the age of AI and big data. On Sunday, Iran unveiled the Simorgh supercomputer, which will deliver.... Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

Leading Solution Providers

Contributors

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

GTC21: Nvidia Launches cuQuantum; Dips a Toe in Quantum Computing

April 13, 2021

Yesterday Nvidia officially dipped a toe into quantum computing with the launch of cuQuantum SDK, a development platform for simulating quantum circuits on GPU-accelerated systems. As Nvidia CEO Jensen Huang emphasized in his keynote, Nvidia doesn’t plan to build... Read more…

Microsoft to Provide World’s Most Powerful Weather & Climate Supercomputer for UK’s Met Office

April 22, 2021

More than 14 months ago, the UK government announced plans to invest £1.2 billion ($1.56 billion) into weather and climate supercomputing, including procuremen Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

Q&A with Jim Keller, CTO of Tenstorrent, and an HPCwire Person to Watch in 2021

April 22, 2021

As part of our HPCwire Person to Watch series, we are happy to present our interview with Jim Keller, president and chief technology officer of Tenstorrent. One of the top chip architects of our time, Keller has had an impactful career. Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

Senate Debate on Bill to Remake NSF – the Endless Frontier Act – Begins

May 18, 2021

The U.S. Senate today opened floor debate on the Endless Frontier Act which seeks to remake and expand the National Science Foundation by creating a technology Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire