Monday, November 13

I Am HPC: Impact and Future Directions

MODERATOR

Valerie E. Taylor

Division Director and Distinguished Fellow, Argonne National Laboratory

Panelists

Enobong (Anna) Branch

Senior Vice President for Equity, Rutgers University

Mateo Valero

Director, Barcelona Supercomputing Center

Katherine (Kathy) Yelick

Robert S. Pepper Distinguished Professor of Electrical Engineering and Computer Sciences; Vice Chancellor for Research at University of California, Berkeley, and Senior Faculty Scientist, Lawrence Berkeley National Laboratory (LBNL)

Frank Indiviglio

Chief Technology Officer, National Oceanic & Atmospheric Administration (NOAA)

Valerie E. Taylor

Valerie E. Taylor is Director of the Mathematics and Computer Science Division and a Distinguished Fellow at Argonne National Laboratory. Prior to her role at Argonne, she served as Senior Associate Dean of Academic Affairs in the College of Engineering along with being a Regents Professor and the Royce E. Wisenbaker Professor in the Department of Computer Science and Engineering (CSE) at Texas A&M University. Taylor also contributed significantly to the academic community during her 11-year tenure in Northwestern University’s Electrical Engineering and Computer Science (EECS) Department. Her research specialization centers on HPC with a primary focus on performance analysis and modeling of parallel scientific applications. 

Currently, Taylor’s research endeavors revolve around enhancing energy efficiency in parallel scientific applications and microelectronics. In addition to her distinguished research career, Taylor is an IEEE, ACM, and American Association for the Advancement of Science (AAAS) Fellow. She extends her impact as the Chief Executive Officer and President of Center for Minorities and People with Disabilities in Information Technology (CMD-IT), a non-profit organization dedicated to creating and delivering innovative programs that champion and advance underrepresented communities in the tech industry, benefiting society at large.

In her role as Senior Vice President for Equity at Rutgers University, Enobong Branch is a prominent advocate for equity and inclusion. Branch leads efforts to embed equity principles in research, education, and public engagement across the university, emphasizing diversity’s role in achieving excellence. Previously, Branch integrated diversity into the strategic plan, conducted a campus climate survey, and championed diversity infrastructure at the University of Massachusetts Amherst. She also led the National Science Foundation-funded ADVANCE program, promoting institutional transformation and gender and racial equity among faculty.

As a sociology professor, Branch explores labor and work, delving into racial and gender inequality’s historical and contemporary aspects. Her publications, including Work in Black and White and Opportunity Denied: Limiting Black Women to Devalued Work, have had a significant impact in addressing historical inequities. Branch’s expertise in diversity in academia extends to advisory roles in professional societies and organizations, including the National Academy of Sciences. She holds a doctorate in sociology from the University at Albany and a B.S. in biology from Howard University.

Mateo Valero

Mateo Valero is a renowned leader in HPC, serving as the founding director of the Barcelona Supercomputing Center and affiliated with the University of Catalonia (UPC). With a prolific career, he has authored 700 papers, played key roles in over 300 international conferences, and delivered more than 600 invited talks. His accolades include prestigious awards such as the Eckert-Mauchly Award, Seymour Cray Award, and Charles Babbage Award from IEEE.

Valero’s influence extends globally with recognition such as HPCWire’s “Readers’ Choice Awards” for his leadership and contributions to European HPC independence. He holds honorary doctorates from 10 universities, is an IEEE and ACM Fellow, and has received high honors from the Mexican government. His remarkable legacy also includes being named a “Favourite Son” of his hometown, Alfamén (Spain), where a public school was named after him in 2006.

Kathy Yelick’s expertise lies in HPC, focusing on programming languages, compilers, parallel algorithms, and automatic performance tuning. Currently, she leads the ExaBiome Project for scalable microbial data analysis tools and co-leads the Berkeley Benchmarking and Optimization (BeBOP) group. In addition to her current role as the Robert S. Pepper Distinguished Professor of EECS, Yelick’s distinguished career has included roles as the Associate Dean for Research at UC Berkeley, Associate Laboratory Director for Computing Sciences at LBNL, and leadership of the National Energy Research Scientific Computing Center. She is a member of the National Academy of Engineering and AAAS, as well as an ACM and AAAS Fellow.

Frank Indiviglio

Frank Indiviglio became NOAA’s Chief Technology Officer in April of 2022. In this role, Frank leads the development of technology and innovation programs across NOAA. Frank also leads research and evaluation of emerging technologies to strengthen NOAA’s ability to accelerate its ability to conduct its mission through technology.

Frank has more than 20 years of experience in technology leadership spanning high performance computing, cloud services, machine learning, media, infrastructure, and enterprise services. Before becoming the CTO, Frank was the Deputy Director of the High Performance Computing and Communications Program at NOAA. He has also held roles at NOAA’s Geophysical Fluid Dynamics Laboratory, Columbia and Cornell Universities, and within industry at HPTi and Silicon Graphics.

Frank holds a Bachelor of Science degree in Information Systems from Drexel University.

Katerina Antypas

Launching the National AI Research Resource (NAIRR) Through a Coordinated Pilot

Katerina (Katie) Antypas is a distinguished SC23 panelist with a rich background in HPC and cyberinfrastructure. Antypas is the Director of the National Science Foundation’s (NSF) Office of Advanced Cyberinfrastructure, where she oversees the development and deployment of advanced computing, data, and networking research resources for the research and education community. Before joining NSF, Antypas had a 17-year tenure at the National Energy Research Scientific Computing (NERSC) Center at Lawrence Berkeley National Laboratory, where she held various roles, including Deputy Division Director, Project Director, Director of Hardware and Integration for the Exascale Computing Project, Data Department Head, and User Services Group Lead. She holds an M.S. in computer science from the University of Chicago and a bachelor’s degree in physics from Wellesley College.

Antypas’s presentation, Launching the National AI Research Resource (NAIRR) Through a Coordinated Pilot, will elaborate on the NAIRR Task Force (NAIRRTF), established by the National Artificial Intelligence AI Initiative Act of 2020, to investigate the creation of a national infrastructure for AI research. After 18 months, the task force produced a report outlining a vision for the NAIRR as a widely accessible U.S. national infrastructure comprising a set of federated resources, including HPC, cloud computing, testbeds, software, datasets, and accompanying user support and educational/training materials. The overarching goal of the envisioned NAIRR is to strengthen and democratize the U.S. AI innovation ecosystem by spurring innovation, increasing the diversity of talent in AI, improving U.S. AI research and development (R&D) capacity, and advancing trustworthy AI. This includes increasing research opportunities in critical areas such as testing and evaluation, bias mitigation, AI safety, and privacy.

Today, a U.S. Government Interagency Working Group, led by the White House Office of Science and Technology and NSF, is in the process of deploying a NAIRR pilot to demonstrate the value, capabilities, and impact of its concept. This initiative aims to reach broad communities, expose technical issues early, and test the proposed NAIRR governance structure. Antypas’s invited talk will describe the latest status and plans for the NAIRR pilot.

Jairo Panetta

Is There Room for HPC in Developing Countries?

Jairo Panetta, currently a volunteer collaborative professor at the Instituto Tecnológico de Aeronáutica (Aeronautics Institute of Technology, Computer Sciences Division [ITA]), is a distinguished academic with extensive expertise in computer science. He specializes in computer architecture, parallel processing, HPC, scientific computing, and supercomputing.

Panetta’s practical approach to computer science has advanced daily production programs, and he has been actively involved in HPC software development, education, and industry applications. His contributions span various sectors, including oil and gas, weather forecasting, and finance. Panetta’s career includes the creation of HPC groups and development of cutting-edge HPC software. Even after retirement, he continues to teach graduate classes on parallel processing at ITA, reflecting his dedication to advancing the field.

Panetta’s presentation, Is There Room for HPC in Developing Countries?, will offer a comprehensive exploration of the policies, key players, and institutions that have shaped HPC in Brazil over the past four decades. Drawing from his rich experience, he will illuminate the HPC landscape in academia, professional societies, and industries. Panetta will also highlight the proactive involvement of federal and state governments in Brazil to foster HPC development. His presentation seeks to provide valuable insights and strategies for other developing countries looking to invest in HPC technology. This invited talk affords an opportunity to gain knowledge and inspiration from an expert who has spent decades at the forefront of the HPC evolution in Brazil.

Amanda randles

Unlocking Potential: The Role of HPC in Computational Medicine

Amanda Randles is the Alfred Winborne Mordecai and Victoria Stover Mordecai Associate Professor of Biomedical Sciences and Biomedical Engineering at Duke University. She holds courtesy appointments in multiple departments, including Mechanical Engineering and Material Science, Computer Science, and Mathematics, and is affiliated with the Duke Cancer Institute. Her research at the intersection of HPC, machine learning, and physics-based modeling focuses on diagnosing and treating diseases, spanning from cardiovascular issues to cancer. Her impressive list of accolades includes the National Institutes of Health Pioneer Award, IEEE TCHPC Early Career Award for HPC, NSF CAREER Award, ACM Grace Hopper Award, ACM/IEEE-CS George Michael Memorial High-Performance Computing Fellowship, and recognition in the World Economic Forum Young Scientist List and MIT Technology Review’s Top 35 Innovators Under 35. She’s also a Fellow of the National Academy of Inventors. Randles earned her doctorate in applied physics from Harvard University, with prior master’s and bachelor’s degrees from Harvard and Duke University, respectively. Before her doctoral studies, she worked as a software engineer at IBM on the Blue Gene supercomputing team. Randles’ substantial contributions encompass over 80 peer-reviewed papers, more than 100 granted U.S. patents, and an additional 100 pending patent applications.

Randles’ presentation, Unlocking Potential: The Role of HPC in Computational Medicine, will delve into the field of computational medicine, focusing on the utilization of vascular digital twins that is revolutionizing healthcare. These models, patient-specific replicas of vascular systems, offer non-invasive solutions for precise blood flow assessments, treatment guidance, and surgical planning. Medical professionals benefit from enhanced expertise through virtual surgery and intervention evaluations. In her invited talk, Randles will address challenges in using digital twins, including data size, computational demands, and integration with wearable data and virtual reality interfaces. She will highlight how HPC emerges as the linchpin, powering the construction and deployment of these intricate models and ultimately unlocking their transformative potential in healthcare. Randles’ presentation will explore HPC’s vital role in advancing personalized computational medicine, shedding light on the complexities and exciting possibilities of this innovative technology.

Sandra Diaz Pier

Boosting Neuroscience Research with High-Performance Computing Infrastructure

Sandra Diaz Pier, a prominent scientist and researcher at Forschungszentrum Juelich in Germany, has made significant contributions in the fields of computational neuroscience at the intersection with high performance computing. Her research has focused on the efficient simulation of brain dynamics at different scales, structural plasticity, optimization and learning in biologically inspired neural networks.

Diaz Pier’s presentation, Boosting Neuroscience Research with High-Performance Computing Infrastructure, will offer a comprehensive overview of the evolving relationship between neuroscience and HPC. Her talk will highlight notable scientific achievements and breakthroughs made possible through this partnership of domain science and computing, showcasing the transformative power of HPC in advancing the understanding of the brain. Moreover, this invited talk will explore the future role of neuroscience in shaping technology co-design with a significant emphasis on its societal impact. Personal anecdotes and international experiences will add depth to her presentation, providing a well-rounded perspective on the convergence of neuroscience and HPC and its potential implications for both scientific progress and society at large.

Denice Ward hood

The Intriguing Interplay Between Culture, Identity, and the World of Computing

Denice Ward Hood is Interim Director of the Center for Culturally Responsive Evaluation and Assessment (CREA) at the University of Illinois Urbana-Champaign (UIUC). She is also a Teaching Professor at UIUC College of Education’s Department of Education Policy, Organization & Leadership, where she has held other leadership roles, such as Director of Online Programs and Director of Graduate Studies.

With more than 30 years of experience, Hood’s expertise lies in program evaluation, learning outcomes assessment, and social science research, predominantly within higher education. Her comprehensive teaching portfolio covers a range of courses, from college student development to race, gender, and sexuality issues.

Beyond her academic role, Hood actively engages in the professional community. She has served as the President of the Arizona Educational Research Association and played a crucial role as Program Chair for the Multiethnic Issues in Evaluation Topical Interest Group (TIG) at the American Evaluation Association. Her expertise extends to program evaluations at local and state levels, as well as her involvement in shaping evaluation plans for numerous NSF grants. She holds a bachelor’s degree in psychology from Northern Illinois University, a master’s degree in educational psychology from the University of Illinois Urbana-Champaign, and a doctorate in educational psychology from Arizona State University.

In her invited talk, Hood will delve into the intriguing interplay between culture, identity, and the world of computing. She will explore the profound impact of individual identities and workplace culture on creativity, innovation, and productivity in the field of technology. From computing’s inclusive origins to its contemporary exclusivity, she will examine the societal implications and role of lived experiences in shaping the technological landscape. She will open a discussion to consider the rich repository of diverse perspectives as a source of remarkable ideas, echoing the call for a more equitable and meaningful future in computing.

Lois Curfman McInnes

Broadening Participation in HPC: Together We Can Change the World

Lois Curfman McInnes is a Distinguished Fellow with Argonne National Laboratory‘s Mathematics and Computer Science Division. Her primary focus is on high-performance computational science, particularly in scalable numerical libraries, and fostering collaborative efforts to build productive and sustainable software ecosystems. She also serves as the Deputy Director of Software Technology for the Exascale Computing Project (ECP) and is a leader in the ECP Broadening Participation Initiative. She also co-leads the Interoperable Design of Extreme-scale Application Software (IDEAS) productivity project, aimed at enhancing software productivity and sustainability to advance scientific research.

McInnes has a doctorate in applied mathematics from the University of Virginia and a B.S. in mathematics and physics from Muhlenberg College. Her early work focused on developing scalable nonlinear solvers in the Portable, Extensible Toolkit for Scientific Computation (PETSc), which provides the foundation of a variety of high-performance scientific applications. She received the 2015 SIAM/ACM Prize in Computational Science & Engineering and a 2009 R&D 100 Award as part of the core PETSc development group, and the E.O. Lawrence Award in 2011 for outstanding contributions in R&D supporting the U.S. Department of Energy (DOE) and its missions.

She is a member of ACM, IEEE, and SIAM. She also is a SIAM Fellow and serves on SIAM Council. McInnes chairs the SIAM Activity Group on Supercomputing (2022-2023), which launched the Supercomputing Spotlights webinar series that features short presentations highlighting the impact and successes of HPC worldwide.

McInnes’ presentation, Broadening Participation in HPC: Together We Can Change the World, will discuss how achievements in HPC, encompassing computational and data-enabled science, analytics, learning, and AI, drive advancements in science and technology globally. She will touch on the ways in which collaborative efforts, such as those within the ECP, are pushing boundaries in various scientific and engineering disciplines through cutting-edge software technologies tailored for exascale computer architectures.

To tackle the pressing challenges of the 21st century effectively, McInnes will emphasize the need for the HPC community to expand its workforce to better reflect the broader society, including underrepresented groups. Her invited talk will provide an overview of inclusive workforce initiatives across the HPC community and provide opportunities for engagement, including a detailed overview of the ECP Broadening Participation Initiative.

McInnes also will highlight the inaugural Intro to HPC Bootcamp, held in August 2023, which featured HPC energy justice projects in collaboration with advanced computing facilities at Argonne, Lawrence Berkeley, and Oak Ridge National Laboratories with the Sustainable Horizons Institute, DOE’s Office of Economic Impact and Diversity, and academic partners.

Carol Scarlett

Quantum Computing Infrastructure and Advances

Carol Scarlett is a distinguished experimental nuclear and particle physicist and a respected faculty member at Florida A&M University (FAMU). Her research portfolio spans a wide range of topics, including areas such as nuclear decay rates; searches for exotic particles, including dark matter; and recent contributions to the development of a Quantum Random Number Generation (QRNG) on a chip-scaled device. Scarlett’s unique ability to bridge disciplines can be attributed to her undergraduate background in electrical engineering and her master’s work in nuclear physics. She brings this interdisciplinary perspective to her research, making her a valuable asset to the scientific community. Furthermore, Scarlett’s expertise extends to her role as a former Fellow of the Chain Reaction Innovations (CRI) program at Argonne National Laboratory. During her time at CRI, she concentrated on leveraging her patented technique to design on-chip QRNG, further showcasing her commitment to pushing innovation in the field.

Scarlett’s presentation, Quantum Computing Infrastructure and Advances, will take a deeper look at quantum computing. As the expansion of various quantum computing platforms continues to surpass earlier predictions, Scarlett will elaborate on the shifts toward achieving the ultimate goal: making quantum computing accessible to end users. She will touch on the myriad of challenging engineering problems that need to be overcome throughout this endeavor.

Scarlett’s invited talk will provide valuable insights into the ongoing work that is paving the way for the future of quantum computing by examining current microelectronics research efforts aimed at driving advancements in quantum computing hardware. Her focus will be on the upcoming generations of quantum computers that face critical issues, such as scalable, fault-tolerant hardware; microdevices with lower energy consumption; standardization of hardware across different qubit types; and development of devices less susceptible to environmental noise. In parallel, she will speak about the urgent need for software platforms to evolve to make quantum hardware accessible while emphasizing the need for workforce development initiatives to equip users with the necessary skills and knowledge.

Fred Chong

Physics-Aware, Full-Stack Quantum Software Optimizations

Fred Chong is the Seymour Goodman Professor in the Department of Computer Science at the University of Chicago and Chief Scientist for Quantum Software at Infleqtion.

As the Lead Principal Investigator for the EPiQC Project (Enabling Practical-scale Quantum Computing), an NSF Expedition in Computing, Chong is at the forefront of advancing practical quantum computing solutions. He plays a vital role in shaping the future of quantum computing policy as a member of the National Quantum Initiative Advisory Committee (NQIAC), offering guidance to the U.S. President via the National Quantum Initiative Program.

Chong earned his doctorate from Massachusetts Institute of Technology in 1996 and has held esteemed faculty positions, including Chancellor’s Fellow at the University of California, Davis from 1997-2005, and Professor of Computer Science, Director of Computer Engineering, and Director of the Greenscale Center for Energy-Efficient Computing at University of California, Santa Barbara from 2005-2015. He is a recognized IEEE Fellow and has received several prestigious awards throughout his career, including the NSF CAREER award, Intel Outstanding Researcher Award, and 13 Best Paper awards. 

Chong’s presentation, Physics-Aware, Full-Stack Quantum Software Optimizations, will explore how quantum software can be a potent tool, significantly accelerating the timeline for achieving utility-scale results with quantum hardware. He will delve into several critical research directions aimed at unlocking practical quantum advantages. One major focus will be on physics-aware, cross-layer optimizations, which continue to yield significant efficiencies, enabling applications to harness quantum resources more effectively. Addressing errors through software-directed mitigation is vital for increasing gate depths and maintaining acceptable output fidelity. Pulse-level optimizations and the use of specialized native gates are also pivotal in this context. Furthermore, Chong will emphasize the importance of hybrid computations, where quantum hardware complements high-performance classical resources such as specialized accelerators. Effectively partitioning computations between these classical and quantum resources is essential for supporting real-world applications. Deep compiler optimization and classical simulation of Clifford and near-Clifford circuits are additional areas of significance in enhancing quantum computation efficiency.