Since 1987 - Covering the Fastest Computers in the World and the People Who Run Them
Since 1987 - Covering the Fastest Computers in the World and the People Who Run Them
June 23, 2022 — The silicon-based computer chips that power our modern devices require vast amounts of energy to operate. Despite ever-improving computing efficiency, information technology is projected to consume around 25% of all primary energy produced by 2030. Researchers in the microelectronics and materials sciences communities are seeking ways to sustainably manage the global need for computing power.
The holy grail for reducing this digital demand is to develop microelectronics that operate at much lower voltages, which would require less energy and is a primary goal of efforts to move beyond today’s state-of-the-art CMOS (complementary metal-oxide semiconductor) devices.
Non-silicon materials with enticing properties for memory and logic devices exist; but their common bulk form still requires large voltages to manipulate, making them incompatible with modern electronics. Designing thin-film alternatives that not only perform well at low operating voltages but can also be packed into microelectronic devices remains a challenge.
Now, a team of researchers at Lawrence Berkeley National Laboratory (Berkeley Lab) and UC Berkeley have identified one energy-efficient route – by synthesizing a thin-layer version of a well-known material whose properties are exactly what’s needed for next-generation devices.
First discovered more than 80 years ago, barium titanate (BaTiO3) found use in various capacitors for electronic circuits, ultrasonic generators, transducers, and even sonar.
Crystals of the material respond quickly to a small electric field, flip-flopping the orientation of the charged atoms that make up the material in a reversible but permanent manner even if the applied field is removed. This provides a way to switch between the proverbial “0” and “1” states in logic and memory storage devices – but still requires voltages larger than 1,000 millivolts (mV) for doing so.
Seeking to harness these properties for use in microchips, the Berkeley Lab-led team developed a pathway for creating films of BaTiO3 just 25 nanometers thin – less than a thousandth of a human hair’s width – whose orientation of charged atoms, or polarization, switches as quickly and efficiently as in the bulk version.
“We’ve known about BaTiO3 for the better part of a century and we’ve known how to make thin films of this material for over 40 years. But until now, nobody could make a film that could get close to the structure or performance that could be achieved in bulk,” said Lane Martin, a faculty scientist in the Materials Sciences Division (MSD) at Berkeley Lab and professor of materials science and engineering at UC Berkeley who led the work.
Historically, synthesis attempts have resulted in films that contain higher concentrations of “defects” – points where the structure differs from an idealized version of the material – as compared to bulk versions. Such a high concentration of defects negatively impacts the performance of thin films. Martin and colleagues developed an approach to growing the films that limits those defects. The findings were published in the journal Nature Materials.
To understand what it takes to produce the best, low-defect BaTiO3 thin films, the researchers turned to a process called pulsed-laser deposition. Firing a powerful beam of an ultraviolet laser light onto a ceramic target of BaTiO3 causes the material to transform into a plasma, which then transmits atoms from the target onto a surface to grow the film. “It’s a versatile tool where we can tweak a lot of knobs in the film’s growth and see which are most important for controlling the properties,” said Martin.
Martin and his colleagues showed that their method could achieve precise control over the deposited film’s structure, chemistry, thickness, and interfaces with metal electrodes. By chopping each deposited sample in half and looking at its structure atom by atom using tools at the National Center for Electron Microscopy at Berkeley Lab’s Molecular Foundry, the researchers revealed a version that precisely mimicked an extremely thin slice of the bulk.
“It’s fun to think that we can take these classic materials that we thought we knew everything about, and flip them on their head with new approaches to making and characterizing them,” said Martin.
Finally, by placing a film of BaTiO3 in between two metal layers, Martin and his team created tiny capacitors – the electronic components that rapidly store and release energy in a circuit. Applying voltages of 100 mV or less and measuring the current that emerges showed that the film’s polarization switched within two billionths of a second and could potentially be faster – competitive with what it takes for today’s computers to access memory or perform calculations.
The work follows the bigger goal of creating materials with small switching voltages, and examining how interfaces with the metal components necessary for devices impact such materials. “This is a good early victory in our pursuit of low-power electronics that go beyond what is possible with silicon-based electronics today,” said Martin.
“Unlike our new devices, the capacitors used in chips today don’t hold their data unless you keep applying a voltage,” said Martin. And current technologies generally work at 500 to 600 mV, while a thin film version could work at 50 to 100 mV or less. Together, these measurements demonstrate a successful optimization of voltage and polarization robustness – which tend to be a trade-off, especially in thin materials.
Next, the team plans to shrink the material down even thinner to make it compatible with real devices in computers and study how it behaves at those tiny dimensions. At the same time, they will work with collaborators at companies such as Intel Corp. to test the feasibility in first-generation electronic devices. “If you could make each logic operation in a computer a million times more efficient, think how much energy you save. That’s why we’re doing this,” said Martin.
This research was supported by the U.S. Department of Energy (DOE) Office of Science. The Molecular Foundry is a DOE Office of Science user facility at Berkeley Lab.
Source: Rachel Berkowitz, Lawrence Berkeley National Laboratory
Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!
You may recall that efforts proposed in 2020 to remake the National Science Foundation (Endless Frontier Act) have since expanded and morphed into two gigantic bills, the America COMPETES Act in the U.S. House of Representatives and the U.S. Innovation and Competition Act in the U.S. Senate. So far, efforts to reconcile the two pieces of legislation have snagged and recent reports... Read more…
Cerebras Systems makes the largest chip in the world, but is already thinking about its upcoming AI chips as learning models continue to grow at breakneck speed. The company’s latest Wafer Scale Engine chip is indeed the size of a wafer, and is made using TSMC’s 7nm process. The next chip will pack in more cores to handle the fast-growing compute needs of AI, said Andrew Feldman, CEO of Cerebras Systems. Read more…
Additional details of the architecture of the exascale El Capitan supercomputer were disclosed today by Lawrence Livermore National Laboratory’s (LLNL) Terri Quinn in a presentation delivered to the 79th HPC User Forum Read more…
The insatiable need to compress time to insights from massive and complex datasets is fueling the demand for quantum computing integration into high performance computing (HPC) environments. Such an integration would allow enterprises to accelerate and optimize current HPC applications and processes by simulating and emulating them on today’s noisy... Read more…
HPCwire presents our interview with Jeff McVeigh, vice president and general manager, Super Compute Group, Intel Corporation, and an HPCwire 2022 Person to Watch. McVeigh shares Intel's plans for the year ahead, his pers Read more…
This post was contributed by Ampersand’s Jeffrey Enos, Senior Machine Learning Engineer, Daniel Gerlanc, Senior Director for Data Science, and Brandon Willard, Data Science Lead. Read more…
There are strict rules governing financial institutions with a number of global regulatory groups publishing financial compliance requirements. Financial institutions face many challenges and legal responsibilities for risk management, compliance violations, and failure to catch financial fraud. Read more…
The long-troubled, hotly anticipated MareNostrum 5 supercomputer finally has a vendor: Atos, which will be supplying a system that includes both Nvidia and Intel CPUs and GPUs across multiple partitions. The newly reimag Read more…
You may recall that efforts proposed in 2020 to remake the National Science Foundation (Endless Frontier Act) have since expanded and morphed into two gigantic bills, the America COMPETES Act in the U.S. House of Representatives and the U.S. Innovation and Competition Act in the U.S. Senate. So far, efforts to reconcile the two pieces of legislation have snagged and recent reports... Read more…
Cerebras Systems makes the largest chip in the world, but is already thinking about its upcoming AI chips as learning models continue to grow at breakneck speed. The company’s latest Wafer Scale Engine chip is indeed the size of a wafer, and is made using TSMC’s 7nm process. The next chip will pack in more cores to handle the fast-growing compute needs of AI, said Andrew Feldman, CEO of Cerebras Systems. Read more…
Additional details of the architecture of the exascale El Capitan supercomputer were disclosed today by Lawrence Livermore National Laboratory’s (LLNL) Terri Read more…
The insatiable need to compress time to insights from massive and complex datasets is fueling the demand for quantum computing integration into high performance computing (HPC) environments. Such an integration would allow enterprises to accelerate and optimize current HPC applications and processes by simulating and emulating them on today’s noisy... Read more…
HPCwire presents our interview with Jeff McVeigh, vice president and general manager, Super Compute Group, Intel Corporation, and an HPCwire 2022 Person to Watc Read more…
The long-troubled, hotly anticipated MareNostrum 5 supercomputer finally has a vendor: Atos, which will be supplying a system that includes both Nvidia and Inte Read more…
Starting today, D-Wave Systems is providing access to a 500-plus-qubit prototype of its forthcoming 7000-qubit Advantage2 quantum annealing computer, which is d Read more…
AMD is getting personal with chips as it sets sail to make products more to the liking of its customers. The chipmaker detailed a modular chip future in which customers can mix and match non-AMD processors in a custom chip package. "We are focused on making it easier to implement chips with more flexibility," said Mark Papermaster, chief technology officer at AMD during the analyst day meeting late last week. Read more…
Getting a glimpse into Nvidia’s R&D has become a regular feature of the spring GTC conference with Bill Dally, chief scientist and senior vice president of research, providing an overview of Nvidia’s R&D organization and a few details on current priorities. This year, Dally focused mostly on AI tools that Nvidia is both developing and using in-house to improve... Read more…
Intel has shared more details on a new interconnect that is the foundation of the company’s long-term plan for x86, Arm and RISC-V architectures to co-exist in a single chip package. The semiconductor company is taking a modular approach to chip design with the option for customers to cram computing blocks such as CPUs, GPUs and AI accelerators inside a single chip package. Read more…
In April 2018, the U.S. Department of Energy announced plans to procure a trio of exascale supercomputers at a total cost of up to $1.8 billion dollars. Over the ensuing four years, many announcements were made, many deadlines were missed, and a pandemic threw the world into disarray. Now, at long last, HPE and Oak Ridge National Laboratory (ORNL) have announced that the first of those... Read more…
AMD/Xilinx has released an improved version of its VCK5000 AI inferencing card along with a series of competitive benchmarks aimed directly at Nvidia’s GPU line. AMD says the new VCK5000 has 3x better performance than earlier versions and delivers 2x TCO over Nvidia T4. AMD also showed favorable benchmarks against several Nvidia GPUs, claiming its VCK5000 achieved... Read more…
The 59th installment of the Top500 list, issued today from ISC 2022 in Hamburg, Germany, officially marks a new era in supercomputing with the debut of the first-ever exascale system on the list. Frontier, deployed at the Department of Energy’s Oak Ridge National Laboratory, achieved 1.102 exaflops in its fastest High Performance Linpack run, which was completed... Read more…
The first-ever appearance of a previously undetectable quantum excitation known as the axial Higgs mode – exciting in its own right – also holds promise for developing and manipulating higher temperature quantum materials... Read more…
The battle for datacenter dominance keeps getting hotter. Today, Nvidia kicked off its spring GTC event with new silicon, new software and a new supercomputer. Speaking from a virtual environment in the Nvidia Omniverse 3D collaboration and simulation platform, CEO Jensen Huang introduced the new Hopper GPU architecture and the H100 GPU... Read more…
PsiQuantum, founded in 2016 by four researchers with roots at Bristol University, Stanford University, and York University, is one of a few quantum computing startups that’s kept a moderately low PR profile. (That’s if you disregard the roughly $700 million in funding it has attracted.) The main reason is PsiQuantum has eschewed the clamorous public chase for... Read more…
Intel reiterated it is well on its way to merging its roadmap of high-performance CPUs and GPUs as it shifts over to newer manufacturing processes and packaging technologies in the coming years. The company is merging the CPU and GPU lineups into a chip (codenamed Falcon Shores) which Intel has dubbed an XPU. Falcon Shores... Read more…
AMD is getting personal with chips as it sets sail to make products more to the liking of its customers. The chipmaker detailed a modular chip future in which customers can mix and match non-AMD processors in a custom chip package. "We are focused on making it easier to implement chips with more flexibility," said Mark Papermaster, chief technology officer at AMD during the analyst day meeting late last week. Read more…
Just a couple of weeks ago, the Indian government promised that it had five HPC systems in the final stages of installation and would launch nine new supercomputers this year. Now, it appears to be making good on that promise: the country’s National Supercomputing Mission (NSM) has announced the deployment of “PARAM Ganga” petascale supercomputer at Indian Institute of Technology (IIT)... Read more…
MLCommons today released its latest MLPerf inferencing results, with another strong showing by Nvidia accelerators inside a diverse array of systems. Roughly fo Read more…
Additional details of the architecture of the exascale El Capitan supercomputer were disclosed today by Lawrence Livermore National Laboratory’s (LLNL) Terri Read more…
The long-troubled, hotly anticipated MareNostrum 5 supercomputer finally has a vendor: Atos, which will be supplying a system that includes both Nvidia and Inte Read more…
A new industry consortium aims to establish a die-to-die interconnect standard – Universal Chiplet Interconnect Express (UCIe) – in support of an open chipl Read more…
Supercomputing has been indispensable throughout the Covid-19 pandemic, from modeling the virus and its spread to designing vaccines and therapeutics. But, desp Read more…
© 2022 HPCwire. All Rights Reserved. A Tabor Communications Publication
HPCwire is a registered trademark of Tabor Communications, Inc. Use of this site is governed by our Terms of Use and Privacy Policy.
Reproduction in whole or in part in any form or medium without express written permission of Tabor Communications, Inc. is prohibited.