Quantum Monte Carlo at Exascale Could Be Key to Finding New Semiconductor Materials

By Linda Barney

September 27, 2021

Researchers are urgently trying to identify possible materials to replace silicon-based semiconductors. The processing power in modern computers continues to increase even as the size of the silicon on which components are placed shrinks. The silicon chip size continues to decrease from 14 nanometers (nm) to 10nm, to 7nm and even smaller. The laws of physics suggest that there is a limit to the number of transistors that can fit on silicon-based chips. As this limit is reached, processing gains will decrease and there will be electron interactions between the components on the chip.

This is one of the research challenges being pursued by scientists preparing for the upcoming Intel-HPE exascale supercomputer, Aurora, which will be housed at the U.S. Department of Energy’s (DOE) Argonne National Laboratory. Supported by the Argonne Leadership Computing Facility’s (ALCF) Aurora Early Science Program, an Argonne team is performing Quantum Monte Carlo research to help locate a new material that could replace silicon-based semiconductors. ALCF computational scientist Anouar Benali is the Primary Investigator on the project.

Anouar Benali

Benali explains, “The search for the next semiconductor material will require high-performance computers (HPC) at exascale. Simulations can only be done on very small prototypes of semiconductor materials with current supercomputers. Using an exascale supercomputer, such as the future Aurora system, will allow us to expand simulations of more materials to help find a viable semiconductor material. Our team is also working with ALCF and the Intel Parallel Computing Center in modifying codes to run on the Aurora supercomputer. We have already seen a 20x processing speedup on the ALCF’s Theta supercomputer, which will allow us to run larger and more realistic compounds in our search for a new semiconductor material.”

What is Quantum Monte Carlo?

The Quantum Monte Carlo (QMC) calculation is a quantum calculation that provides some of the most accurate solutions to quantum mechanical problems. QMC provides theoretical predictions for many problems at the forefront of research—from materials science to complex biological systems. The main power of the QMC method is that instead of trying to solve analytically the Schrödinger equation, describing and predicting all interactions in nature, QMC generates millions of random solutions. QMC then accepts all the solutions that solve the equation or rejects all the ones that don’t to save research time and improve prediction accuracy.


Semiconductor Quantum Chemistry Research

Predicting the properties of materials, or designing materials based on desired properties, is one of the most important goals of material science simulations. Benali states, “Most of the phenomena driving these properties occur at very small scales that are ruled by the laws of quantum mechanics.” For example, knowing if a material will be a good semiconductor or fine tuning the composition of a material to generate the perfect semiconductor requires solving the many-body problem of interacting particles in a quantum system. Researchers often use density functional theory (DFT) equations for organic molecules to determine the properties of a many-electron system. DFT simulations are less computationally expensive than QMC, but the predictions are not as accurate.

Benali’s team developed the open-source simulation code QMCPACK, software which contains Monte Carlo algorithms and uses the Schrödinger equation in calculations. Benali explains, “By solving the Schrödinger equation using statistical methods, large and complex systems can be studied to unprecedented accuracy—including systems where other electronic structure methods have difficulty. The Schrödinger equation can predict all behavior of almost everything in the universe. However, solving the equation for a system as small as a hydrogen molecule is impossible without a significant number of approximations using computer simulations. When simulating properties of more complex materials such as semiconductors, the number of approximations becomes significant, with a tradeoff of a loss of accuracy.”

“The team can significantly reduce the number of quantum approximations that can be simulated using an exascale system such as the future Aurora supercomputer. Using an exascale system will allow us to run calculations on larger systems as well as increase the accuracy of the simulation result to help identify a new semiconductor material to replace silicon-based semiconductors,” states Benali.

Scientific Case Study

The team’s research involves searching for a substitute for silicon (Si) complementary metal-oxide-semiconductor (CMOS) based computing materials. In locating a suitable material to replace silicon, the team must address a fundamental materials problem that current can leak through a hafnium(IV) oxide (HfO2) gate dielectric. Hafnium is used in optical coatings, and as a high-κ dielectric in Dynamic Random-Access Memory (DRAM) capacitors and in advanced metal-oxide-semiconductor devices. Researchers have found evidence that impurities like nitrogen and fluorine are able to reduce leakage currents. However, current computational studies are limited to DFT which do not provide the necessary accuracy to test this theory.

The team is using QMCPACK software to study the energetics of point defects near a HfO2 interface. QMCPACK is funded by the DOE Exascale Comupting Project. Due to the large number of electrons in these simulations, calculations can only be made possible by the large aggregate memory and performance offered by an exascale system such as the future Aurora supercomputer. Benali states, “Our research seeks to prove that adding the right amount of the right impurities can enable the properties we are looking for in the next generation of semiconductors. But mostly, our research is trying to demonstrate that with enough computer power, quantum simulations can be fully predictive and provide significant support to experimental research.”

HfO2 semiconductor (Hf in blue, Oxygen in red) between Pt connectors (white) Courtesy of Dr. Benali, Argonne National Laboratory.

Reducing Calculations with Future Exascale Supercomputers

Increased compute power enables the research team to use QMC to significantly reduce the number of approximations in the resolution of the Schrödinger equation. “The more solutions we try, the more accurate our results become. The greatest value of using the Monte Carlo approach resides in the fact that each random solution is independent from another, meaning that the resolution can be distributed on as many processing units as are available,” indicates Benali.

How Long Does Semiconductor Research Take?
If the accuracy of the answer requires simulating one million random solutions and simulating one solution takes one second on one processing unit, the team’s simulation will end in one million seconds (277 days). However, because the evaluations of the random solutions are independent, on a machine like Argonne’s Theta supercomputer, the evaluation takes about 3.55 seconds. In a more realistic simulation, some operations of initialization and collection of data cannot be parallelized and increase the cost of the computation, but overall the method remains extremely parallel.


Updating QMCPACK Software: Exascale Systems Require a New Approach to Software

The open-source QMCPACK code used by the team in their in the semiconductor research is maintained by Argonne National Laboratory, Oak Ridge National Laboratory, Lawrence Livermore National Laboratory and several universities. Benali is one of the co-owners and co-developers of QMCPACK. Intel has been working with this group since 2016 and provides access to multiple experimental computer systems along with tools such as the Intel Math Kernel Library (MKL) and Intel compilers. The ALCF team maintains the software and code stacks on test machines. In addition, ALCF staff aids researchers in creating or modifying code so that it performs the desired chemistry functions and is designed to run on the future Aurora exascale system.

Coding software for an exascale system requires a new approach to software design and coding. Hardware on an exascale system is not a heterogenous system. In an exascale system, there is not one single Central Processing Unit (CPU) but many components such as CPUs, Graphic Processing Units (GPUs), Field Programmable Gate Arrays (FPGAs), or expanded memory that performs different tasks. The team is breaking the program functions into various modules that perform calculations for a component such as a CPU, multiple CPUs, a GPU as well as calculations based on memory. Coding for an exascale system requires that the programmer is more aware of the varied architecture of the system when writing code. In addition, the team’s semiconductor research can do millions of simulations that are running completely separately. So modifying QMCPACK code to optimize running on GPUs is a priority.

Challenges for Future Quantum Materials Research

“In our current research, our team uses quantum material science simulations to try to predict the behavior and structure of materials to help locate a suitable material to replace silicon in semiconductors. Our team uses QMCPACK on large scale HPC in our quantum chemistry research. “With the advent of the exascale era, a machine like Aurora will allow simulating more realistic and complex materials, replicating what experimentalists do in-situ. Such leaps will allow to orders of magnitude speed up in the pace of discoveries” indicates Benali.

The ALCF is a DOE Office of Science User Facility. Primary support for QMCPACK is via the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences and Engineering Division, as part of the Computational Materials Sciences Program and Center for Predictive Simulation of Functional Materials, and also the Exascale Computing Project (17-SC-20-SC), a collaborative effort of the U.S. Department of Energy Office of Science and the National Nuclear Security Administration.

References

  • “QMCPACK: Advances in the development, efficiency, and application of auxiliary field and real-space variational and diffusion quantum Monte Carlo” , P. R.C. Kent, A Annaberdiyev, A Benali et. al. J. Chem. Phys. 152 (17), 174105 (2020)
  • ”Defect energetics of cubic hafnia from quantum Monte Carlo simulations”, Raghuveer Chimata, Hyeondeok Shin, Anouar Benali, and Olle Heinonen, Phys. Rev. Materials 3, 075005 – (2019)
  • “Zirconia and hafnia polymorphs: Ground-state structural properties from diffusion Monte Carlo” H. Shin, A. Benali, Y. Luo, E. Crabb, A. Lopez-Bezanilla, L.E. Ratcliff, A. M. Jokisaari and O. Heinonen, Phys Rev. Mat 2 (7), 075001 (2018)
  • “QMCPACK : An open source ab initio Quantum Monte Carlo package for the electronic structure of atoms, molecules, and solids”, J. Kim, A. Baczewski , T. D. Beaudet, A. Benali , M C. Bennett, M. A Berrill , N. S Blun, M. Casula, D. M Ceperley, S. Chiesa, B. K Clark, R. C Clay III, K. T Delaney, M. Dewing, K. P Esler, H. Hao, O. Heinonen, P. R C Kent, J. T Krogel, I. Kylänpää, Y. Wai Li, M G. Lopez, Y. Luo, R. M Martin, A. Mathuriya, J. McMinis, C. A Melton, L. Mitas, M. A Morales, E. Neuscamman, W. D Parker, S. D Pineda Flores, N. A Romero, B. M Rubenstein, J. A R Shea, H. Shin, L. Shulenburger, A. Tillack, J. P Towsend, N. M. Tubman, B. Van Der Goetz, J. E Vincent, S. Zhang, L. Zhao, Y. Yang. J. Phys. Cond. Mat. 30 195901 (2018)

Linda Barney is the founder and owner of Barney and Associates, a technical/marketing writing, training, and web design firm in Beaverton, Ore.

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!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues 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 power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion 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…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen 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…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. 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…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a 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…

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…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

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…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

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…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire