Berkeley Lab Computing Sciences

It’s , a special day dedicated to promoting and appreciating the mathematical constant τ (tau), equal to 2π, or approximately 6.28. As enthusiasts of mathematics and its applications, Berkeley Lab Computing Sciences Area staff celebrated with pizza, cookies, trivia, and hula hooping!

Learn more about Tau Day: https://tauday.com/
cc: U.S. Department of Energy

Thank you to everyone who came to the Open Chiplet Economy Experience Center at Berkeley Lab today!
In the post-Moore’s Law era of computing, specialized architectures like chiplets are redefining computing and providing new avenues for advancing supercomputing speed and energy efficiency. This year represents a significant milestone for the open chiplet economy: the start of a transition from documents and slides into the realm of tangible and real products.
Co-hosted by Berkeley Lab Computing Sciences and the Open Compute Project Foundation, today’s event provided a select collection of technology demonstrations covering products and prototypes emerging from different corners of the emerging chiplet market that will impact the development of specialized ASICs for HPC.
Listen to Berkeley Lab's John Shalf discuss the "Rise of the Killer Chiplets" on the OrionX.net podcast earlier this year: https://bit.ly/chipletshpc

cc: U.S. Department of Energy

Check out the latest episode of Big Brains Podcast podcast by The University of Chicago where Berkeley Lab's Michael Wehner lends his expertise on the escalation of hurricanes that are becoming so extreme we might need a category 6.
Listen here: https://bit.ly/4c1hYui

cc: U.S. Department of Energy Earth & Environmental Sciences Area

Storm Warning: Why Hurricanes Are Growing Beyond Measure, with Michael Wehner | Big Brains We all know that extreme weather events like hurricanes are getting worse due to climate change, but what scientists would really like to know is: By how much worse exactly? This year a team of researchers argued that hurricanes have become so much more extreme due to climate change that we need to....

In a world striving for sustainability, understanding the hidden half of a living plant – the roots – is crucial.
As part of an investigation to boost agricultural yields and develop crops resilient to climate change, Berkeley Lab researchers developed RhizoNet. This computational tool harnesses the power of AI to transform how we study plant roots and root behavior under various environmental conditions.
Learn more: https://bit.ly/RhizoNet

cc: U.S. Department of Energy

Yesterday, we officially welcomed more than 140 visiting faculty and student researchers to Berkeley Lab’s Computing Sciences Area (CSA) for summer 2024. For more than a decade, this program has offered high school, undergraduate, and graduate students in science and engineering the chance to gain research experience working with staff, mentors, and peers in Berkeley Lab's CSA, including the Applied Math and Computational Research (AMCR) and Scientific Data (SciData) Divisions, Energy Sciences Network (ESnet), and the National Energy Research Scientific Computing Center (NERSC).
SciData Division Director Ana Kupresanin and AMCR Division Director Stefan Wild hosted the kickoff event yesterday, along with AMCR Scientist and 2024 CSA Summer Student Program Chair Osni Marques.
Learn more about what we have in store for our program participants this summer: https://bit.ly/CSAsp24
Learn more about our mentorship programs: https://bit.ly/LBNLcsSR


cc: U.S. Department of Energy Sustainable Horizons Institute

Numerical analysis is a cornerstone of scientific research, often working behind the scenes to drive significant advancements. In a groundbreaking paper published in Physical Review X, mathematicians from Berkeley Lab and UC Berkeley showcased its pivotal role in quantum chemistry. They utilized coupled-cluster theory to develop a more efficient method for identifying finite-size errors in condensed matter systems. This innovative approach not only enhances our understanding of material properties but also helps predict these properties accurately before the materials move into the manufacturing phase. This leap forward illustrates how numerical analysis can transform theoretical insights into practical applications, paving the way for future technological innovations.
https://bit.ly/powerofnumericalanalysis
U.S. Department of Energy

Did you know that Berkeley Lab researchers recently unveiled an interpretable AI model and research that sheds light on a captivating discovery: the brain is a dual marvel? It acts as a dynamic system, seamlessly orchestrating our perception, thoughts, and actions while simultaneously functioning as a high-powered computing engine, deftly processing sensory, cognitive, and behavioral information.

Their findings were published in PLOS Computational Biology. Read more: https://bit.ly/PLOScb

cc: U.S. Department of Energy National Institutes of Health (NIH)

Last month, Applied Mathematician Phil Colella’s colleagues, collaborators, mentors, and mentees convened at Berkeley Lab’s Wang Hall to celebrate an illustrious career that has spanned more than four decades and includes formative contributions to the U.S. Department of Energy's applied mathematics and computing programs, as well as the fields of applied math, computational science, and engineering.
In addition to his tenure in Berkeley Lab Computing Sciences, Colella was a staff scientist at Lawrence Livermore National Laboratory and a Berkeley Mechanical Engineering professor.
Read more: https://bit.ly/PColella

cc: Berkeley Engineering UC Berkeley University of California Society for Industrial and Applied Mathematics ACM - Association for Computing Machinery

Experimental breakthroughs via Berkeley Lab's Advanced Quantum Testbed (AQT) user program: An interdisciplinary team at Lawrence Livermore National Laboratory, University of Washington, UC Berkeley, and Berkeley Lab proposed a hybrid algorithm for simulating the (real-time) dynamics of quantum mechanical systems of particles. U.S. Department of Energy Office of Science.
The researchers successfully used the hybrid computing scheme to simulate the scattering of two neutrons. This opens a path to computing nuclear reaction rates that are difficult or impossible to measure in a laboratory. These include reaction rates that play a role in astrophysics and national security. The hybrid scheme will also aid in simulating the properties of other quantum mechanical systems. For example, it could help researchers study the scattering of electrons with quantized atomic vibrations known as phonons, a process that underlies superconductivity.
Read more: https://bit.ly/hybridAQT

cc:

Novel hybrid scheme speeds the way to simulating nuclear reactions on quantum computers The nuclear reactions that power the stars and forge the elements emerge from the interactions of the quantum mechanical particles, protons and neutrons. Explaining these processes is one of the most challenging unsolved problems in computational physics.

From keynote sessions to tutorials, BOFs, and workshops, Berkeley Lab is bringing its HPC expertise to the ISC High Performance 2024 Conference in Hamburg, Germany, May 12-16. To kick things off, on May 13 Kathy Yelick will deliver the opening keynote, “Beyond Exascale Computing.” And on May 15, John Shalf will be part of a keynote session on “Reinventing HPC with Specialized Architectures and New Applications Workflows.” For more information and a complete list of Berkeley Lab’s participation at ISC, go to https://cs.lbl.gov/news-media/news/2024/berkeley-lab-researchers-dive-into-hpcs-future-at-isc-2024/

With the exascale era underway, the community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that have occurred about once a decade.
With this in mind, Kathy Yelick, Berkeley Lab’s Senior Advisor on Computing and UC Berkeley’s Vice Chancellor of Research, will be giving the keynote talk “Beyond Exascale Computing” at this year’s International Supercomputing Conference (ISC24) in Hamburg, Germany.
Ahead of her talk, she spoke to Intersect360 Research Senior Analyst Steve Conway about some of these post-exascale challenges: https://bit.ly/KYbeyondexa via HPCwire

cc: U.S. Department of Energy Berkeley Engineering

Kathy Yelick on Post-Exascale Challenges With the exascale era underway, the HPC community is already turning its attention to zettascale computing, the next of the 1,000-fold performance leaps that have occurred about once a decade. […]

Congratulations to Irfan Siddiqi on his election to the American Academy of Arts & Sciences!
Siddiqi is the director of Berkeley Lab’s Advanced Quantum Testbed, which develops and operates full-stack quantum computing platforms based on superconducting qubits. He’s also a faculty scientist in the Lab’s Applied Mathematics and Computational Research Division (AMCR) and head of the UC Berkeley Physics Department.
Learn more: https://bit.ly/AAASIrfan

__________________________________________
cc: U.S. Department of Energy

In California's 2022-2023 winter season, nine atmospheric rivers caused severe flooding, landslides, and power outages, the longest continuous weather condition of this type in 70 years.
Recently, scientists at Berkeley Lab used machine learning to study these weather systems and found that more intense atmospheric rivers are more likely to occur successively within a short period.
Learn more: https://bit.ly/ARca

cc: U.S. Department of Energy Earth & Environmental Sciences Area

Unraveling the mysteries of consecutive atmospheric river events In California’s 2022-2023 winter season, the state faced nine atmospheric rivers (ARs) that led to extreme flooding, landslides, and power outages – the longest duration of continuous AR conditions in the past 70 years. Scientists recently conducted a study using machine learning to better under...

Did you know that researchers from Berkeley Lab, NERSC, San Francisco State University, and Case Western Reserve University recently published two papers introducing new methods of data storage and analysis to make quantum computing more practical and exploring how visualization helps in understanding quantum computing?
"This work represents significant strides in understanding and harnessing current quantum devices for data encoding, processing, and visualization. These contributions build on our previous efforts to highlight the ongoing exploration and potential of quantum technologies in shaping scientific data analysis and visualization,” said Talita Perciano, a research scientist in Berkeley Lab's Scientific Data Division and the leader of this effort.
Read more: https://bit.ly/QISTlbl

cc: U.S. Department of Energy

🎉 🎉 Congratulations to Berkeley Lab's David Brown!
He’s being honored with a 2024 Society for Industrial and Applied Mathematics (SIAM) Prize for Distinguished Service to the Profession in recognition of his decades of commitment to enriching the computational science community: https://bit.ly/SIAMdb
This prize will be awarded at the SIAM Annual Meeting (AN24), which will take place July 8-12, 2024, in Spokane, WA. Established in 1985, this prize is awarded annually for contributions to the advancement of applied mathematics on the national or international level.

CC: U.S. Department of Energy Sustainable Horizons Institute Krell Institute Los Alamos National Laboratory Lawrence Livermore National Laboratory

David Brown Honored with SIAM Prize for Distinguished Service to the Profession Berkeley Lab’s David Brown has been honored with a 2024 SIAM Prize for Distinguished Service to the Profession for his decades of commitment to enriching the computational science community.

🎉 Congratulations to Berkeley Lab’s Chao Yang! He's been named a 2024 Society for Industrial and Applied Mathematics fellow for contributions to the development of novel algorithms and software for eigenvalue problems, as well as applications of algorithms and software to solve challenging scientific problems: https://bit.ly/SIAMcy

Yang's work has helped chemists and material scientists gain insights into the electronic, structural, and optical properties of molecules comprising tens of thousands of atoms. His algorithms have been used to optimize particle accelerator designs, investigate the nuclear structures of isotopes, and better understand protein structures. And many of his algorithms for electronic structure calculation have also been incorporated into widely used application software such as Quantum-Espresso, NWChem, Q-Chem, and PARSEC.

cc: U.S. Department of Energy

AMCR’s Chao Yang Named 2024 SIAM Fellow Berkeley Lab’s Chao Yang has been named a 2024 SIAM fellow for contributions to the development of novel algorithms and software for eigenvalue problems, as well as applications of algorithms and software to solve challenging scientific problems.

🎉 Congratulations to Berkeley Lab Scientific Data Division's Fernando Perez!
Project Jupyter, a non-profit, open-source project born out of the IPython Project that he started, has been named a winner of The White House Office of Science and Technology Policy (OSTP)'s Year of Open Science Recognition Challenge.
Read the OSTP Announcement: https://bit.ly/ProjectJupyterOSTP
Read about Perez's road from IPython to Project Jupyter: https://bit.ly/IPythonPJ

cc: U.S. Department of Energy UC Berkeley

Project Jupyter: A Computer Code that Transformed Science A computer code (Project Jupyter) co-developed by Berkeley Lab's Fernando Perez and embraced by the global science community over two decades has been hailed by Nature Magazine as one of “ten computer codes that transformed science.”

Berkeley Lab Research Scientist Nan Ding shares her path to HPC in this spotlight from the SC Conference!
cc: U.S. Department of Energy

I feel HPC is a super cool thing! I enjoy developing methodologies to guide and accelerate science. Currently, I’m focused on innovations in using performance models to understand the potential and pitfalls of system architectures and am excited to present the work I’ve been undertaking in evaluating disaggregated memory systems at the forthcoming MEMCON conference happening this month right in the heart of Silicon Valley! My career has been focused on improving the performance of HPC solutions - and I’ve been lucky to have been able to help fine-tune everything from HPC applications, to presenting a visual performance model for GPUs and future system architectures.

Disaggregated memory is a promising approach that addresses the limitations of traditional memory architectures by enabling memory to be decoupled from compute nodes and shared across a data center. Cloud platforms have deployed such systems to improve overall system memory utilization, but performance can vary across workloads. HPC is crucial in scientific and engineering applications, where HPC machines also face the issue of underutilized memory. As a result, improving system memory utilization while understanding workload performance is essential for HPC operators. Therefore, learning the potential of a disaggregated memory system before deployment is a critical step. I apply my methodology to analyze thirteen diverse workloads, including AI training, data analysis, genomics, protein, fusion, atomic nuclei, and traditional HPC bookends, and demonstrate that intra-rack memory disaggregation would meet the application's memory requirement and provide enough remote memory bandwidth.

My name is Nan Ding, and I am a Research Scientist at Lawrence Berkeley National Laboratory.

🎉🎉🎉 Congratulations to Berkeley Lab’s Katie Antypas! She’s been named one of HPCwire’s People to Watch in 2024: https://bit.ly/HPCwireKA
Last year, Antypas temporarily stepped away from her role as Deputy Director of NERSC to become the Director of the National Science Foundation (NSF)’s Office of Advanced Infrastructure, where she’s leading the launch and deployment of the National AI Research Resource (NAIRR) pilot effort, an urgent and critical initiative for the future of AI innovation, with ten other federal agencies and 25 industry and non-profit partners.
“My advice to young people is to take the hard classes and worry less about your grades. Nobody ever said in retrospect, ‘you know, I really wish I understood less math.’ Take the hard classes. You will have the skills you learn with you for the rest of your life. Also, don’t be afraid to switch fields or directions. I came to Computer Science quite late. I didn’t take a CS course until my senior year in college, but I loved it and adjusted directions,” she said.

cc: U.S. Department of Energy

Quantum hardware is experiencing a boon leading to more chip variety and configurations with higher fidelities. While ultimately, this will translate to a boon for the entire field of quantum computing, it presents a software design problem by placing more of the overall burden of realizing end-to-end quantum applications on the software stacks, specifically the quantum compiler.
The Berkeley Quantum Synthesis Toolkit (BQSKit) is a powerful and portable quantum compiler framework with a proven ability to alleviate this issue and translate recent hardware successes up to the algorithm level.
One of BQSKit’s developers, Berkeley Lab’s Ed Younis, will give a virtual talk about this tool this Sunday, March 17, 2024, from 9 - 11 a.m. PDT.
Learn more about the talk and register at https://bit.ly/BQSKit_ey

cc: U.S. Department of Energy

Compiling Resource-Efficient Programs with Numerical Instantiation Quantum hardware is experiencing a boon leading to more chip variety and configurations with higher fidelities.

Did you know that Berkeley Lab's Scientific Data Division researchers have developed a novel AI-based method for diagnosing input/output (I/O) performance bottlenecks in high performance computing (HPC)? Their new tool, AIIO (Artificial Intelligence for I/O), automatically identifies I/O bottlenecks at the job level and offers potential solutions.
In experiments run at NERSC, this approach demonstrated that real applications and even unseen applications (those not used in the training model) can use these diagnostic results to improve I/O performance on HPC systems.
Learn more: https://bit.ly/AIIO

cc: U.S. Department of Energy Scientific Data Management Group at LBNL

On the path to modern scientific discovery, successful research now requires a partnership between domain scientists and computational experts who harness computing power, build a digital roadmap, and troubleshoot technical issues to move the mission forward.
To help support software engineers in academia and science, a national advocacy organization – known as the US Research Software Engineer Association (US-RSE) – has made it its mission to mount a “community-driven effort focused on the increasingly important role” of research software engineers. And Berkeley Lab’s Keith Beattie has been recently elected to US-RSE's steering committee: https://bit.ly/US-RSElbl

cc: U.S. Department of Energy

When it comes to scientific software, user experience is often an afterthought. This oversight can impact product accessibility, design, functionality, and usability. Fortunately, the UX community for scientific software is growing and has begun to address these challenges with new tools like the Berkeley Lab-developed STRUDEL (Scientific sofTware Research for User experience, Design, Engagement and Learning) Design System.
The STRUDEL team is actively seeking users from across the science community to test drive this first version of the Design System and provide feedback that will enhance its performance, utility, and usability going forward.
Toward this end, they’re hosting a seminar and hackathon this month: https://bit.ly/STRUDELhackathon.
*March 8, 2024, 11 a.m.–12 p.m. PT.
“How to Build a Science UI: Getting Started with the STRUDEL Design System.”
Register: https://lbnl.zoom.us/meeting/register/tJYrcu6hqDwvGNJeB1Mw8DGI04BM7bIdxp1K #/registration
* March 19, 2024, All-Day Hackathon in Berkeley, where participants will be able to prototype a new UI using the Design System while receiving expert support from the STRUDEL team.
Register by March 11: https://www.surveymonkey.com/r/1ststrudelhackathon.

cc: U.S. Department of Energy

STRUDEL to Hold Hands-on Hackathon at Berkeley Lab on March 19 STRUDEL, an emerging open-source science project at Berkeley Lab, is hosting a webinar and a hackathon in March to introduce software teams to a new user experience/user interface design tool that addresses the challenges of creating usable scientific software and puts user engagement at the forefro...

Last week, quantum researchers across Berkeley Lab and UC Berkeley hosted a delegation of quantum networking experts from Japan, including Dr. S***a Nagayama, who is leading the Quantum Network effort for Japan’s Moonshot Goal 6, and Keio University professor Dr. Rod Van Meter.
During the two-day visit, they met with QUANT-NET lead Inder Monga; Berkeley Lab ATAP Division’s Thomas Schenkel; UC Berkeley Professor Alp Sipahigil; Quantum Systems Accelerator Deputy Director Bert de Jong; Advanced Quantum Testbed's Head of Hardware Kasra Nowrouzi, and Berkeley Lab Computing Sciences Area Associate Lab Director Jonathan Carter, who gave an overview of Quantum Information Science at Berkeley Lab.

cc: University of California

Congratulations to Berkeley Lab Affiliate Giulia Guidi! She was just honored with a Society for Industrial and Applied Mathematics (SIAM) Activity Group on Supercomputing Early Career Prize, a recognition bestowed biannually to an individual in their early career for outstanding research contributions in algorithms research and development for parallel scientific and engineering computing.
"It’s such an honor to receive this prize. It’s a very big recognition. I’m just at the beginning of my career, so it is really exciting that the work I’ve been doing has been recognized by the community,” said Guidi.
Learn more: https://bit.ly/GGuidi
cc: U.S. Department of Energy Cornell University UC Berkeley Berkeley Engineering UC Berkeley Association of Women in EECS

How can machine learning enhance the prognosis and understanding of traumatic brain injury?
A team of researchers from Berkeley Lab, UCSF, Medical College of Wisconsin, and UC Berkeley – in conjunction with the TRACK-TBI (Transforming Research and Clinical Knowledge in Traumatic Brain Injury) collaboration – are investigating, and they've recently detailed some findings in Nature Scientific Reports: https://bit.ly/TrackTBI_LBNL

cc: U.S. Department of Energy University of California Nature Portfolio

Machine Learning Opens New Doors in Traumatic Brain Injury Research In a paper published in Nature Scientific Reports, researchers from Berkeley Lab, UCSF, the Medical College of Wisconsin, and UC Berkeley – in conjunction with the TRACK-TBI collaboration – describe how an innovative machine learning approach can enhance the prognosis and understanding of trauma...

Mark your calendars! 🗓
Berkeley Lab's Lavanya Ramakrishnan will lead a webinar on Wednesday, February 21, 2024, from 9-10 a.m. Pacific Time to discuss STRUDEL (Scientific sofTware Research for User experience, Design, Engagement, and Learning), a project that is bringing UX/UI methods to scientific software design. Learn more: https://bit.ly/STRUDEL221 | Register: https://bit.ly/4bINGgw

A related webinar, “How to Build a Science UI: Getting Started with the STRUDEL Design System,” will take place on March 8, 11 a.m.–12 p.m. Pacific time; for more information and to register, go to https://go.lbl.gov/strudel-hackathon-signup.

cc: U.S. Department of Energy Sloan Foundation Public Understanding of Science & Technology

Did you know that Berkeley Lab's Marcus Noack and Mark Risser have recently proposed a significant advancement in the area of machine learning and data science that promises significant computational improvements: the enhancement of exact Gaussian Processes for large datasets, significantly improving data analysis capabilities for samples even beyond 5 million data points?
Learn more: https://bit.ly/MNMRgp via Scientia

cc: U.S. Department of Energy

As increasing ocean temperatures contribute to more intense and destructive hurricanes, climate scientists Michael Wehner (Berkeley Lab) and James Kossin (First Street Foundation) wondered whether the open-ended Category 5 on the Saffir-Simpson Windscale is sufficient to communicate the risk of hurricane damage in a warming climate.
So they investigated and detailed their extensive research in a new article published in the Proceedings of the National Academy of Sciences (PNAS), where they also introduce a hypothetical Category 6 to hurricanes, which would encompass storms with wind speeds greater than 192 mph.
Learn more: https://bit.ly/Cat6Storms

cc: U.S. Department of Energy

In a Warming World, Climate Scientists Consider Category 6 Hurricanes For more than 50 years, the National Hurricane Center has used the Saffir-Simpson Windscale to communicate the risk of property damage; it labels a hurricane on a scale from Category 1 (wind speeds between 74 - 95 mph) to Category 5 (wind speeds of 158 mph or greater). But as increasing ocean temper...

Did you know there are several different paths for students and academic faculty to work with researchers in Berkeley Lab Computing Sciences? Learn more about our CS Mentorship programs: https://bit.ly/LBNLcsSR.

“This experience has given me the opportunity to diversify my skill set and develop my proficiency in programming languages. I began to understand how programmers approach problem-solving. The frequent testing of code was surprising to me at first, but I’ve come to realize its value, and I hope to bring this mindset with me into future work, computational or otherwise.
Berkeley Lab’s commitment to recruiting diverse individuals and forging collaborations between people from different fields was very important to me. The best learning experiences for me have been those with my colleagues from varied backgrounds. The diversity of thought and skill is beneficial to any research project.”
-Emma Groetecke, University of Kentucky Senior, spoke of her internship in Berkeley Lab’s Computer Languages and Systems Software (CLaSS) Group.
She spent the summer of 2023 working on a project that adapted existing force field development methods to suit gold’s unique chemistry with the hope of accurately predicting gold binding to proteins of interest as part of the Sustainable Research Pathways (SRP) program.

cc: U.S. Department of Energy Berkeley Lab K-12 STEM Sustainable Horizons Institute