IEEE Releases Expansive 2018 Roadmap for Devices and Systems

By John Russell

July 24, 2019

IEEE yesterday released the 2018 update to its International Roadmap for Devices and System (IRDS). It’s part history text, part crystal ball, and as IEEE emphasizes, a key reference for the entire electronics value chain. You may remember this report got its start in ~1998 as the International Technology Roadmap for Semiconductors (ITRS) when the focus was squarely on semiconductor technology. It morphed into the IRDS report in 2017 as the decline in Moore’s Law highlighted the rise of many other factors driving computer performance. IEEE and its collaborators responded by expanding the breadth of the report.

As noted in the announcement: “The updated IRDS includes new information on cryogenic electronics and quantum information processing, added benchmarks for applications, and supplemental information and metrics from the More Moore team. In addition, there’s a newly released summary from Beyond CMOS, and updates for emerging devices, outside systems connectivity technology, factory integration (including smart manufacturing and security topics), metrology, and yield enhancement. Market drivers for medical devices and a new automotive market drivers report are also included.”

Let’s say IRDS is more than light summer reading.

On balance, historical information seems most plentiful and many of the projected trends are familiar; nevertheless there’s plenty to dig into. (Check out HPCwire’s coverage of the final ITRS report, Transistors Won’t Shrink Beyond 2021, Says Final ITRS Report) Below is IEEE’s description of the new report:

“The IRDS is a set of predictions that serves as the successor to the ITRS. The intent is to provide a clear outline to simplify academic, manufacturing, supply, and research coordination regarding the development of electronic devices and systems.

“The goals of the roadmap are as follows:

  1. To identify key trends related to devices, systems, and all related technologies by generating a roadmap with a 15-year horizon
  2. To determine generic devices’ and systems’ needs, challenges, potential solutions, and opportunities for innovation
  3. To encourage related activities worldwide through collaborative events, such as related IEEE conferences and roadmap workshops

“The shift and evolution of the roadmap from the ITRS to the IRDS has translated to an expanded focus on systems. Emphasis has been placed on architectures and applications that deviate from the traditional paradigm of device->circuit->logic gate->functional block->system.”

Those are big objectives. Sorting through the report’s high points is beyond the scope of this article. A fair amount of discussion is given to IoT and cloud computing. HPC proper receives a bit less attention. Best to carve out some beach time for reading. It is interesting to note that the 15-year horizon forecast on some of its predictions seems tenuous or directional at best – it’s hard to know much about what 2033 will look like. One example is shown below with explanatory notes at the bottom of the article.

Among IRDS’s extensive contents, it posits a number of Grand Challenges. The lengthy executive summary (~38 pages) provides an overview. Here are a few elements from a Grand Challenge labelled More Moore:

LOGIC DEVICE SCALING
Beyond 2022 a transition from FinFET to gate-all-around (GAA) will start and potentially a transition to vertical nanowires devices will be needed when there will be no room left for the gate length scale down due to the limits of fin width scaling (saturating the Lgate scaling to sustain the electrostatics control) and contact width.

FinFET and lateral GAA devices enable a higher drive at unit footprint if fin pitch can be aggressively scaled while increasing the fin height. This increased drive at unit footprint by scaling the fin pitch comes at a trade-off between fringing capacitance between gate and contact and series resistance. This trend in reducing the number of fins while balancing the drive with increased fin height is defined as fin depopulation strategy, which also simultaneously reduces the standard cell height, therefore the overall chip area.

The most difficult challenge for interconnects is the introduction of new materials that meet the wire conductivity requirements and reduce dielectric permittivity. As for the conductivity, the impact of size effects on interconnect structures must be mitigated. Future effective κ requirements preclude the use of a trench etch stop for dual damascene structures.

DRAM AND 3D NAND FLASH MEMORY
Since the DRAM storage capacitor gets physically smaller with scaling, the EOT (equivalent oxide thickness) must scale down sharply to maintain adequate storage capacitance. To scale the EOT, dielectric materials having high relative dielectric constant (κ) will be needed. Therefore metal-insulator-metal (MIM) capacitors have been adopted using high-κ (ZrO2/Al2O/ZrO2) as the capacitor of 40−30 nm half-pitch DRAM. This material evolution and improvement will continue until 20 nm high- performance (HP) and ultra-high-κ (perovskite κ > 50 ~ 100) materials are released. Also, the physical thickness of the high-κ insulator should be scaled down to fit the minimum feature size. Due to that, capacitor 3D structure will be changed from cylinder to pillar shape.

It’s probably best to think of IRDS as a living document that has grown by accretion, jettisoning little through the years. The result, perhaps necessarily, is the new material seems perhaps a bit thin. Its strength is still on the “semiconductor side” but expansion to systems and more is the right direction.

“The IRDS continues to lead as the go-to reference for researchers, developers and technologists around the world by providing a comprehensive overview of the computer and electronics industry’s trajectory,” said IEEE Fellow Thomas M. Conte, co-chair, IEEE Rebooting Computing Initiative, vice-chair of IRDS, and professor, Georgia Institute of Technology in the official announcement. “The updated IRDS builds upon 16 years of projecting technology needs for the evolving semiconductor and computer industries.”

“The IRDS represents a global effort needed for future computing systems covering many different applications. These worldwide roadmapping activities will allow our community to identify and overcome emerging challenges in this field and to speed-up technology innovation that can drive the development of future markets,” said Francis Balestra, member, Governing Board of the SiNANO Institute, director of research, The French National Center for Scientific Research (CNRS) and vice president of Grenoble Institute of Technology, also quoted in the release.

IRDS partners with regional roadmaps in Europe and Japan. “There are memorandums of understanding (MoUs) with the NanoElectronics Roadmap for Europe: Identification and Dissemination (NEREID, Horizon 2020), of the SiNANO Institute in Europe, and with the Systems and Devices Roadmap committee of Japan (SDRJ) of the Japan Society of Applied Physics,” according to IEEE.

IEEE reports “The updated IRDS can be downloaded by visiting the IRDS home page and joining the IRDS Technical Community. The IRDS is an IEEE Standards Association (IEEE SA) Industry Connections (IC) Program sponsored by the IEEE Rebooting Computing (IEEE RC) Initiative, a program of IEEE Future Directions.”

Link to IEEE release: https://www.businesswire.com/news/home/20190723005052/en/IEEE-Update-International-Roadmap-Devices-Systems-IRDSTM

Link to IRDS page: https://irds.ieee.org

[1]Notes for Table ES2:

ORTC: Logic Notes[1] Based on 0.71x reduction per “Node Range” (“Node” = ~2x Mx).
[2] Based on 0.71x Mx reduction per “Generic Node”, or .5x cell; 2x density; beginning 2013/”G1″/40nm.
[3] Defined as distance between metallurgical source/drain junctions

ORTC: DRAM Notes
[1] The definition of DRAM Half pitch has been changed from this edition. Because of 6F2 DRAM cell, BL pitch is no more critical dimension. pitch= (Cell Area/ Call size factor)^0.5.”
Critical dimension for process development, the Minimum half pitch is also introduced. Currently Active area (long rectangle island shape) half pitch is the critical dimension of 6F2 DRAM.
Calculated half pitch is use the following equation “Calculated half pitch= (Cell Area/ Call size factor)^0.5.” Critical dimension for process development, the Minimum half pitch is also introduced.
Currently Active area (long rectangle island shape) half pitch is the critical dimension of 6F2 DRAM.
[11] Cell size factor = a = (DRAM cell size/F2), where F is the DRAM 1⁄2 pitch. The current values of a are 6 from 2009. And a=4 will be predicted in 2021.

ORTC: NAND Flash Notes
[1] 2D NAND strings consist of closely packed polysilicon control gates (the Word Lines) that separate the source and drain of devices with no internal contact within the cell.
Up to now this uncontacted word line pitch is still the tightest in all technologies.
[6] The number of 3D layers is not a unique function, depending on the cell 1/2 pitch and 3D NAND technology architecture chosen. Lower number of 3D layers generally has lower bit cost,
but other factors such as decoding method, speed performance, easier or harder to get yield, also need to be considered.
The number of 3D layers spans a range since the same density product may be achieved by using smaller 1/2 pitch and fewer layers, or larger 1/2 pitch and more layers.

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!

Rockport Networks Launches 300 Gbps Switchless Fabric, Reveals 396-Node Deployment at TACC

October 27, 2021

Rockport Networks emerged from stealth this week with the launch of its 300 Gbps switchless networking architecture focused on the needs of the high-performance computing and the advanced-scale AI market. Early customers Read more…

AWS Adds Gaudi-Powered, ML-Optimized EC2 DL1 Instances, Now in GA

October 27, 2021

As machine learning becomes a dominating use case for local and cloud computing, companies are racing to provide solutions specifically optimized and accelerated for AI applications. Now, Amazon Web Services (AWS) is int Read more…

Fireside Chat with LBNL’s Advanced Quantum Testbed Director

October 26, 2021

Last week, Irfan Siddiqi led a “fireside chat” with a few media and analysts to introduce the Department of Energy’s relatively new Advanced Quantum Testbed (AQT), which is based at Lawrence Berkeley National Labor Read more…

Graphcore Introduces Larger-Than-Ever IPU-Based Pods

October 22, 2021

After launching its second-generation intelligence processing units (IPUs) in 2020, four years after emerging from stealth, Graphcore is now boosting its product line with its largest commercially-available IPU-based sys Read more…

Quantum Chemistry Project to Be Among the First on EuroHPC’s LUMI System

October 22, 2021

Finland’s CSC has just installed the first module of LUMI, a 550-peak petaflops system supported by the European Union’s EuroHPC Joint Undertaking. While LUMI -- pictured in the header -- isn’t slated to complete i Read more…

AWS Solution Channel

Royalty-free stock illustration ID: 577238446

Putting bitrates into perspective

Recently, we talked about the advances NICE DCV has made to push pixels from cloud-hosted desktops or applications over the internet even more efficiently than before. Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

Rockport Networks Launches 300 Gbps Switchless Fabric, Reveals 396-Node Deployment at TACC

October 27, 2021

Rockport Networks emerged from stealth this week with the launch of its 300 Gbps switchless networking architecture focused on the needs of the high-performance Read more…

AWS Adds Gaudi-Powered, ML-Optimized EC2 DL1 Instances, Now in GA

October 27, 2021

As machine learning becomes a dominating use case for local and cloud computing, companies are racing to provide solutions specifically optimized and accelerate Read more…

Fireside Chat with LBNL’s Advanced Quantum Testbed Director

October 26, 2021

Last week, Irfan Siddiqi led a “fireside chat” with a few media and analysts to introduce the Department of Energy’s relatively new Advanced Quantum Testb Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the powe Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Quantum Workforce – NSTC Report Highlights Need for International Talent

October 13, 2021

Attracting and training the needed quantum workforce to fuel the ongoing quantum information sciences (QIS) revolution is a hot topic these days. Last week, the U.S. National Science and Technology Council issued a report – The Role of International Talent in Quantum Information Science... Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

US Closes in on Exascale: Frontier Installation Is Underway

September 29, 2021

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, held by Zoom this week (Sept. 29-30), it was revealed that the Frontier supercomputer is currently being installed at Oak Ridge National Laboratory in Oak Ridge, Tenn. The staff at the Oak Ridge Leadership... Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... 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... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was 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…

Leading Solution Providers

Contributors

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... 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 i 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…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-ap 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…

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…

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…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire