AMD Refreshes Roadmap, Transitions Back to HPC

By Tiffany Trader

May 7, 2015

AMD revealed key elements of its multi-year strategy as part of its 2015 Financial Analyst Day event in New York on Wednesday. Out of the gate, CEO Lisa Su acknowledged the company’s recent challenges, pointing to a weak PC market and market share losses, before turning her attention to the game plan that AMD is counting on to turn its earnings statement from red to green. It’s a game plan that has AMD returning to the high-end server space as it seeks to diversify its revenue base and grow into new markets.

“We are focused on areas that require high performance compute, high performance graphics, visualization technologies, and complex system on chips,” said Su, who kicked off the proceedings, “those are the areas that are uniquely suited to AMD…and we think this represents about a $60+ billion TAM.”

AMD FAD 2015_3 year game plan slide

“Datacenter is probably the single biggest bet that we are making as a company,” she declared. “We have not been competitive the last few years, we will be competitive in the datacenter market.”

Su also spoke about the decision to exit the SeaMicro dense server system business line “for one because microservers were not growing as fast as originally thought and two we really aren’t a systems company, however on the silicon side, very very clearly we are an x86 company, we have tremendous x86 heritage and are absolutely going to invest in high-performance x86.”

With regard to technology, Su said that AMD portfolio decisions will be focused on high-performance cores, immersive technologies, 2.5/3D packaging and software/APIs. This will align with increased x86 investment, focused ARM investment and a simplified CPU roadmap.

Su also offered an appeasement for those wondering why AMD didn’t make these changes sooner.

“On the platform side, to those of you that ask what have you guys been doing for a couple of years. The truth is, it takes a while to really transform both the R&D capabilities, the technologies and the modularity,” she said.

And now…Getting Zen with simplified roadmaps

The highlight of AMD’s revamped technology roadmap is a brand new x86 processor core codenamed “Zen,” touting improved instructions per clock of up to 40 percent over “Excavator” cores. Absent from the lineup, however, is the Skybridge project. Announced last year, the plan to join x86 and ARM together on a common platform was dropped, according to Su, due to customer feedback indicating a desire for x86 and ARM, but not necessarily in socket-compatible factors.

AMD+FAD+2015_Zen

Mark Papermaster, AMD chief technology officer and senior vice president, addressed AMD’s x86 positioning and laid out some of Zen’s specs in anticipation of its 2016 debut. AMD is counting on its new Zen core to drive its re-entry into high-performance desktop and server markets and put it back on a competitive track against arch rival Intel.

“It’s got high-throughput, very efficient design, and a new cache and memory subsystem design to feed this core,” he said, referring to the feature of simultaneous multi-threading (SMT). The performance is the result of doubling down on the previous generation core, Excavator, due out this summer, said Papermaster.

“This wasn’t one silver bullet,” Papermaster continued, “but a number of elements combining to drive the microarchitecture improvement and deliver what I’ve not seen in the industry before, a 40 percent improvement in instructions per clock.”

It’s the core design for the workload of the future and it’s available next year, he added. It’s also a commitment to sustainable innovation, according to Papermaster, who says the company has leapfrogging design teams and is already working on the successor to Zen as it works to establish a family of cores over time.

Papermaster also revealed that AMD’s first custom 64-bit ARM core, “K12” core, is on track for a 2017 sampling. These enterprise-class ARM cores are designed for efficiency and are intended for server and embedded workloads.

AMD FAD 2015_cg roadmap detail

On the graphics side, AMD is readying to launch its high-performance graphics processing unit (GPU) with die stacked High Bandwidth Memory (HBM) using a 2.5D silicon interposer design. This core is optimized for graphics and parallel compute and includes a number of other enhancements (depicted in the slide below). AMD reported that future generations of its high-performance GPUs will be based on FinFET process technology, which will contribute to a doubling of performance-per-watt.

AMD+FAD+2015_Graphics+Leadership+slide

These three essential chip technologies will be the building blocks of AMD’s Enterprise, Embedded and Semi-Custom Business Group (EESC). A new group launched in 2014 as part of AMD’s business unit reorganization, EESC is focused on high-priority markets that will leverage high-performance CPU and GPU cores inside differentiated solutions.

Forrest Norrod, senior vice president and general manager of the business group, referred to the EESC segment as “a principal driver of growth for the last few years and one we think is central to the growth story of AMD going forward.”

Norrod added that these three businesses (enterprise, embedded and semi-custom) share a perspective around the best way to showcase AMD technology.

“In all of these businesses our customers are building products around the technology ingredients that we give to them and bringing differentiated solutions to the end customer that leverage AMD unique IP,” he stated.

“So we really think now of ESSC as being a continuum leveraging technology, customer relationships and the modular design approach at both the chip as well as the systems level.”

Norad went on to share in broad strokes AMD’s datacenter roadmap for the 2016-2017 timeframe, which includes its next-gen x86 Opteron, next-gen ARM, and an APU that we will be tracking closely.

AMD FAD 2015_eesc roadmap

The upcoming next-generation AMD Opteron processors are based on the x86 “Zen” core and target mainstream servers. These x86 Opterons tout high core count with full multi-threading, disruptive memory bandwidth and high native I/O capacity. Norrrod also introduced “the highest performance ARM server GPUs,” powered by AMD’s upcoming “K12” core.

Most relevant for HPC, though, is the new high-performance server APU, a multi-teraflops chip targeting HPC and workstation markets.

“We’re bringing the APU concept fully into the server realm,” Norrod stated. “These are high performance server APUs offering not just high-performance CPU cores and memory but multi-teraflops GPU-capability, providing a level of performance for machine learning, a level of performance for finite element analysis, and a level of performance for memory bandwidth for reverse time migration algorithms that the oil companies use to do reservoir simulation.”

The APU line (APU stands for accelerated processing unit) is an outcome of the Fusion project, which started in 2006 with AMD’s acquisition of the graphics chipset manufacturer ATI. AMD has talked up the potential benefits of tight CPU-GPU integration for HPC workloads in the past, but until now AMD’s APU efforts have primarily been relegated to desktop space.

In January 2012, AMD rebranded the Fusion platform as the Heterogeneous Systems Architecture (HSA) and has in recent months begun championing the CPU+accelerator architecture for a wide range of workloads, including HPC.

AMD says the next-gen server APU stands to deliver massive improvements to vector applications with scale-up graphics performance, HSA enablement, and optimized memory architecture. “We think we’ve got unique and compelling technology that is only possible by wedding together the CPU and world-class GPU and combining them with an open standard HSA software interface,” said Norrod.

What’s not clear at this point is whether the APU’s multi-teraflops will be of the half- single- or double-precision variety, and the workloads that Norrod lists are a mixed bag in that respect (FEA and machine learning, for example). Of course, there is no reason AMD can’t launch variants for each, but it would be hard to claim HPC cred without an FP64-heavy version.

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!

Under The Wire: Nearly HPC News (June 13, 2024)

June 13, 2024

As managing editor of the major global HPC news source, the term "news fire hose" is often mentioned. The analogy is quite correct. In any given week, there are many interesting stories, and only a few ever become headli Read more…

Quantum Tech Sector Hiring Stays Soft

June 13, 2024

New job announcements in the quantum tech sector declined again last month, according to an Quantum Economic Development Consortium (QED-C) report issued last week. “Globally, the number of new, public postings for Qu Read more…

Labs Keep Supercomputers Alive for Ten Years as Vendors Pull Support Early

June 12, 2024

Laboratories are running supercomputers for much longer, beyond the typical lifespan, as vendors prematurely deprecate the hardware and stop providing support. A typical supercomputer lifecycle is about five to six years Read more…

MLPerf Training 4.0 – Nvidia Still King; Power and LLM Fine Tuning Added

June 12, 2024

There are really two stories packaged in the most recent MLPerf  Training 4.0 results, released today. The first, of course, is the results. Nvidia (currently king of accelerated computing) wins again, sweeping all nine Read more…

Highlights from GlobusWorld 2024: The Conference for Reimagining Research IT

June 11, 2024

The Globus user conference, now in its 22nd year, brought together over 180 researchers, system administrators, developers, and IT leaders from 55 top research computing centers, national labs, federal agencies, and univ Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst firm TechInsights. Nvidia's GPU shipments in 2023 grew by more Read more…

Under The Wire: Nearly HPC News (June 13, 2024)

June 13, 2024

As managing editor of the major global HPC news source, the term "news fire hose" is often mentioned. The analogy is quite correct. In any given week, there are Read more…

Labs Keep Supercomputers Alive for Ten Years as Vendors Pull Support Early

June 12, 2024

Laboratories are running supercomputers for much longer, beyond the typical lifespan, as vendors prematurely deprecate the hardware and stop providing support. Read more…

MLPerf Training 4.0 – Nvidia Still King; Power and LLM Fine Tuning Added

June 12, 2024

There are really two stories packaged in the most recent MLPerf  Training 4.0 results, released today. The first, of course, is the results. Nvidia (currently Read more…

Highlights from GlobusWorld 2024: The Conference for Reimagining Research IT

June 11, 2024

The Globus user conference, now in its 22nd year, brought together over 180 researchers, system administrators, developers, and IT leaders from 55 top research Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst fir Read more…

ASC24 Expert Perspective: Dongarra, Hoefler, Yong Lin

June 7, 2024

One of the great things about being at an ASC (Asia Supercomputer Community) cluster competition is getting the chance to interview various industry experts and Read more…

HPC and Climate: Coastal Hurricanes Around the World Are Intensifying Faster

June 6, 2024

Hurricanes are among the world's most destructive natural hazards. Their environment shapes their ability to deliver damage; conditions like warm ocean waters, Read more…

ASC24: The Battle, The Apps, and The Competitors

June 5, 2024

The ASC24 (Asia Supercomputer Community) Student Cluster Competition was one for the ages. More than 350 university teams worked for months in the preliminary competition to earn one of the 25 final competition slots. The winning teams... 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…

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…

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…

Everyone Except Nvidia Forms Ultra Accelerator Link (UALink) Consortium

May 30, 2024

Consider the GPU. An island of SIMD greatness that makes light work of matrix math. Originally designed to rapidly paint dots on a computer monitor, it was then 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…

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…

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…

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

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…

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…

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…

Intel’s Next-gen Falcon Shores Coming Out in Late 2025 

April 30, 2024

It's a long wait for customers hanging on for Intel's next-generation GPU, Falcon Shores, which will be released in late 2025.  "Then we have a rich, a very 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…

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…

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…

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