Stacking Stairs Against the Memory Wall

By Nicole Hemsoth

April 2, 2013

Last year at Intel’s Developer Forum, the chipmaker unfurled a plan it cooked up with Micron to release “real” stacked memory into the world. The concept was too appealing to ignore—a potential solvent to dissolve the parallel DRAM performance and efficiency wall, not to mention backed by well-known vendors.

The need in the market isn’t easy to ignore either. With ever-mounting CPU advancements that promise superior performance, the blame for lousy delivery on those chip promises lies squarely on memory. Finding a way to refine how quickly, not to mention efficiently, memory can feed dashing CPUs is a priority–and its one that several vendors have been eyeing.

While Micron and Intel weren’t the first or only to propose clicking together 3D memory, the news cycle around the technology did heat up. Since then, several vendors have stacked their chips on the betting table, including NVIDIA with its recent Volta announcement.

Not long ago we sat down with former Cray scientist, Steve Scott, who is now CTO for the Tesla division at NVIDIA. During our chat, which went beyond GPU computing, we talked about the memory wall and how approaches like stacked DRAM are still “not ready for primetime.” He noted that the challenges of stacking memory are limited by some pure physics barriers—at least if users are going to see bandwidth in the arena of one terabyte per second, as it claims will be possible with its stacked solution by around 2015 with the rollout of their own spin on the stack, Volta.

Of course, not everyone agrees with the assertion that the approach isn’t ready for primetime, particularly those who have been pushing 3D memory concepts—and getting some sizable industry backing in the process. Count among that crowd veteran semiconductor giant, Micron, which is staking a portion of its future on being the foundational force behind ultra-dense stacked memory with its Hybrid Memory Cube.

Just as with other tech that promises to revolutionize performance and efficiency, stacked DRAM could put a big dent in the network and high performance computing side of the memory wall says Mike Black, who serves chief technology strategist for Micron’s Hybrid Memory Cube (HMC). Black helped spearhead an effort announced today that is set to standardize on this technology and make Micron, Samsung and its partners the rally point for the big takeoff they predict for the technology.

Sloughing off the assertion about the HMC not being ready for real production, the Hybrid Memory Consortium is out to bolster stacked memory via an open standard that a vendor community can build on. Supported by founding partner Samsung and a bevy of developers and adopters including Altera, IBM, ARM, Xilinx and others, the effort extended today to set a fresh global standard to bring the dense-memory approach into motion.

The team behind the standard has been working for close to two years to deliver on the final specifications for HMC interface standards. As Jim Elliot, VP of Memory Planning at Samsung said today, the ideal outcome is that “system designers and manufacturers will be able to get new green memory solutions that outperform other memory options offered today.”

As the group noted:

The achieved specification provides an advanced, short-reach (SR) and ultra short-reach (USR) interconnection across physical layers (PHYs) for applications requiring tightly coupled or close-proximity memory support for FPGAs, ASICs and ASSPs, such as high-performance networking, and test and measurement. The next goal for the consortium is to further advance standards designed to increase data rate speeds for SR from 10, 12.5 and 15Gb up to 28Gb. Speeds for USR interconnections will be driven from 10 up to 15Gb. The next level of specification is projected to gain consortium agreement by the first quarter of 2014.

For Micron and the partners they’ve rallied for the Hybrid Memory Cube Consortium (an effort they launched with Samsung), stacked memory is a current reality–and one that presents a viable production opportunity, not to mention a fifteen-fold boost per cube over standard approaches and stated 70% energy usage reduction per bit than average DDR3 DRAM.

And besides, notes Micron’s Black, there wasn’t a lot of value for a company like Micron to be floating around in niche offering territory. To lead in this still immature space they need the backing of a larger community of users, developers and vendors.

There is little doubt that there’s a wall to be scaled on performance and efficiency fronts when it comes to traditional approaches to DRAM, claims Black. The Hybrid Memory Cube technology that his company has led seeks to scale the wall via a memory architecture that matches a high-speed logic layer with a stack of through-silicon via (TSV) bonded memory die.

Micron points to the reality of this technology, noting that the stacked architecture has been shown to  do everything but wash the dishes by kicking out the slower parallel interfaces of current DDR3s to create shorter, more efficient pathways and using 90% less energy than current RDIMMs via the more dense memory approach.

In an effort to get a production-ready offering, Micron launched a demo platform a couple of years ago to vet out some of the challenges that come with trying to cobble together a heterogeneous package of DRAM and logic devices on the same slice.

Black said that when it came to addressing challenges during that test phase, the biggest hurdle was in the TSV technology itself—more specifically, making it viable and ready for mass production. “There was a lot of physics and learning as we developed from that point,” he noted. “We had to ask what was missing, how small in diameter is it possible to make a TSV, what’s the best conductive material, how to negotiate the radio between the TSV diameter and the aspect ratio of the thickness of the silicon wafer—more learning had to happen as we kept stacking.”

Micron sees a day on the horizon when they’ll be able to stack beyond the four and eight-high DRAM units that are a current possibility. When they release this fall, they’ll be able to offer 160 gigabits of bandwidth—so basically a terabit of bandwidth per cube. Of course, once users start blocking multiple cubes together it’s possible to further increase total bandwidth. By the 2016 point, Black says they’ll likely double the bandwidth they’re able to offer now.

Micron plans to start sampling their stacked DRAM device this fall and will ramp it into production in the first half of next year.

Related Articles

Volta Adds Charge to GPU Roadmap

MetaRAM Launches New Memory Technology

Penguin Pushes Envelope on Compute Density

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!

Data Vortex Users Contemplate the Future of Supercomputing

October 19, 2017

Last month (Sept. 11-12), HPC networking company Data Vortex held its inaugural users group at Pacific Northwest National Laboratory (PNNL) bringing together about 30 participants from industry, government and academia t Read more…

By Tiffany Trader

AI Self-Training Goes Forward at Google DeepMind

October 19, 2017

DeepMind, Google’s AI research organization, announced today in a blog that AlphaGo Zero, the latest evolution of AlphaGo (the first computer program to defeat a Go world champion) trained itself within three days to play Go at a superhuman level (i.e., better than any human) – and to beat the old version of AlphaGo – without leveraging human expertise, data or training. Read more…

By Doug Black

Researchers Scale COSMO Climate Code to 4888 GPUs on Piz Daint

October 17, 2017

Effective global climate simulation, sorely needed to anticipate and cope with global warming, has long been computationally challenging. Two of the major obstacles are the needed resolution and prolonged time to compute Read more…

By John Russell

HPE Extreme Performance Solutions

Transforming Genomic Analytics with HPC-Accelerated Insights

Advancements in the field of genomics are revolutionizing our understanding of human biology, rapidly accelerating the discovery and treatment of genetic diseases, and dramatically improving human health. Read more…

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Cluster Competition coverage has come to its natural home: H Read more…

By Dan Olds

Data Vortex Users Contemplate the Future of Supercomputing

October 19, 2017

Last month (Sept. 11-12), HPC networking company Data Vortex held its inaugural users group at Pacific Northwest National Laboratory (PNNL) bringing together ab Read more…

By Tiffany Trader

AI Self-Training Goes Forward at Google DeepMind

October 19, 2017

DeepMind, Google’s AI research organization, announced today in a blog that AlphaGo Zero, the latest evolution of AlphaGo (the first computer program to defeat a Go world champion) trained itself within three days to play Go at a superhuman level (i.e., better than any human) – and to beat the old version of AlphaGo – without leveraging human expertise, data or training. Read more…

By Doug Black

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Read more…

By Dan Olds

Intel Delivers 17-Qubit Quantum Chip to European Research Partner

October 10, 2017

On Tuesday, Intel delivered a 17-qubit superconducting test chip to research partner QuTech, the quantum research institute of Delft University of Technology (TU Delft) in the Netherlands. The announcement marks a major milestone in the 10-year, $50-million collaborative relationship with TU Delft and TNO, the Dutch Organization for Applied Research, to accelerate advancements in quantum computing. Read more…

By Tiffany Trader

Fujitsu Tapped to Build 37-Petaflops ABCI System for AIST

October 10, 2017

Fujitsu announced today it will build the long-planned AI Bridging Cloud Infrastructure (ABCI) which is set to become the fastest supercomputer system in Japan Read more…

By John Russell

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Intel Debuts Programmable Acceleration Card

October 5, 2017

With a view toward supporting complex, data-intensive applications, such as AI inference, video streaming analytics, database acceleration and genomics, Intel i Read more…

By Doug Black

Reinders: “AVX-512 May Be a Hidden Gem” in Intel Xeon Scalable Processors

June 29, 2017

Imagine if we could use vector processing on something other than just floating point problems.  Today, GPUs and CPUs work tirelessly to accelerate algorithms Read more…

By James Reinders

NERSC Scales Scientific Deep Learning to 15 Petaflops

August 28, 2017

A collaborative effort between Intel, NERSC and Stanford has delivered the first 15-petaflops deep learning software running on HPC platforms and is, according Read more…

By Rob Farber

How ‘Knights Mill’ Gets Its Deep Learning Flops

June 22, 2017

Intel, the subject of much speculation regarding the delayed, rewritten or potentially canceled “Aurora” contract (the Argonne Lab part of the CORAL “ Read more…

By Tiffany Trader

Oracle Layoffs Reportedly Hit SPARC and Solaris Hard

September 7, 2017

Oracle’s latest layoffs have many wondering if this is the end of the line for the SPARC processor and Solaris OS development. As reported by multiple sources Read more…

By John Russell

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Google Releases Deeplearn.js to Further Democratize Machine Learning

August 17, 2017

Spreading the use of machine learning tools is one of the goals of Google’s PAIR (People + AI Research) initiative, which was introduced in early July. Last w Read more…

By John Russell

GlobalFoundries Puts Wind in AMD’s Sails with 12nm FinFET

September 24, 2017

From its annual tech conference last week (Sept. 20), where GlobalFoundries welcomed more than 600 semiconductor professionals (reaching the Santa Clara venue Read more…

By Tiffany Trader

Graphcore Readies Launch of 16nm Colossus-IPU Chip

July 20, 2017

A second $30 million funding round for U.K. AI chip developer Graphcore sets up the company to go to market with its “intelligent processing unit” (IPU) in Read more…

By Tiffany Trader

Leading Solution Providers

Amazon Debuts New AMD-based GPU Instances for Graphics Acceleration

September 12, 2017

Last week Amazon Web Services (AWS) streaming service, AppStream 2.0, introduced a new GPU instance called Graphics Design intended to accelerate graphics. The Read more…

By John Russell

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

EU Funds 20 Million Euro ARM+FPGA Exascale Project

September 7, 2017

At the Barcelona Supercomputer Centre on Wednesday (Sept. 6), 16 partners gathered to launch the EuroEXA project, which invests €20 million over three-and-a-half years into exascale-focused research and development. Led by the Horizon 2020 program, EuroEXA picks up the banner of a triad of partner projects — ExaNeSt, EcoScale and ExaNoDe — building on their work... Read more…

By Tiffany Trader

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Cray Moves to Acquire the Seagate ClusterStor Line

July 28, 2017

This week Cray announced that it is picking up Seagate's ClusterStor HPC storage array business for an undisclosed sum. "In short we're effectively transitioning the bulk of the ClusterStor product line to Cray," said CEO Peter Ungaro. Read more…

By Tiffany Trader

Intel Launches Software Tools to Ease FPGA Programming

September 5, 2017

Field Programmable Gate Arrays (FPGAs) have a reputation for being difficult to program, requiring expertise in specialty languages, like Verilog or VHDL. Easin Read more…

By Tiffany Trader

IBM Advances Web-based Quantum Programming

September 5, 2017

IBM Research is pairing its Jupyter-based Data Science Experience notebook environment with its cloud-based quantum computer, IBM Q, in hopes of encouraging a new class of entrepreneurial user to solve intractable problems that even exceed the capabilities of the best AI systems. Read more…

By Alex Woodie

Intel, NERSC and University Partners Launch New Big Data Center

August 17, 2017

A collaboration between the Department of Energy’s National Energy Research Scientific Computing Center (NERSC), Intel and five Intel Parallel Computing Cente Read more…

By Linda Barney

  • arrow
  • Click Here for More Headlines
  • arrow
Share This