More than HPC. The SC20 Planning Committee chose that theme in early 2019 to emphasize that as a community, we are more than just the bits and bytes that we work with to solve the problems facing us today. Little did we know how important that would become. And when it became apparent that, with everyone’s safety as our first concern, we would have to take SC20 online – I chose to change our location not to “virtual,” but to “Everywhere We Are.” That’s because it’s true in any year, and any place.

Conference Program

As I’ve written here before, we were far enough into the submissions and acceptance process for a number of conference activities that we were faced with a difficult decision about how far to take SC20, regardless of where or how it was held: either reduce acceptances for major components of the conference…or go big, and put on the conference with its customary breadth and depth.

We chose the latter and were the better for it. As a committee, 742 of us worked harder than ever planning for various scenarios, stretching ourselves to the limits of our agility while we learned more about our virtual platform’s capabilities even just a few days out from the event as the platform itself evolved in response to lessons learned from earlier virtual conferences. We were able to loosen pre-recording requirements for our tutorials and workshops which allowed them to be much more interactive than we had initially thought possible. We facilitated live Q&A sessions for most of our events, delivered an abundant Technical Program – 30 Tutorials, 40 Workshops, 95 Papers, 16 Birds of a Feather sessions, 106 Posters, an additional COVID-focused Gordon Bell component, 12 Invited Talks, 10 Panels, 18 State of the Practice Talks , a vibrant Early Career program, and even an Awards Ceremony – and had a full complement of Exhibitor Forum sessions and 285 interactive Exhibits to encourage the kind of discussions we knew many exhibitors and attendees would long for.

Our SCinet team put on an engaging SCinet program despite there being no buildout of the world’s fastest temporary network. It’s a great look at the people behind SCinet, and the future of it as well.

With the Students@SC program, we kept students involved as volunteers, content creators, cluster challenge competitors from a record 19 teams, and as a key component of what makes SC the well-rounded conference it is, thanks to our extraordinarily enthusiastic Students@SC committee.

And we’re able to keep this vast trove of technical, exhibits, Students@SC, and SCinet content online for the next six months – over 600 hours of content created by you, for each other.

Speaking of you – over 7,440 of you attended SC20 from over 115 countries.

Inclusivity and Diversity

We had great inclusivity and diversity activities planned for Atlanta, which would have been a new city for the SC Conference Series – and one of the most diverse destinations yet for the conference, home to a significant number of local historically black colleges and universities. Those plans were curtailed in March but we were still able to grant 41 inclusivity registration scholarships to students, faculty, and staff from minority serving institutions in the United States, Turkey, and Indonesia. Twenty-five of those recipients were new to SC.

Our new HPC in the City hackathon featured 43 student participants and 21 mentors who used HPC resources and methodologies to analyze a wide range of important Atlanta-specific topics, including K-12 broadband access in underserved areas in Atlanta, human trafficking, voter fraud, and COVID-19 impact on the economy of Fulton County, GA.

And Computing4Change returned, sponsored by ACM SIGHPC – allowing 19 undergraduates the chance to use XSEDE resources Jetstream and Stampede2 to work on student-chosen problems including the environment, overall health, violence, and COVID-19 disease spread, mobility, and its impacts upon healthcare.

Keynote and Plenary: Climate Science and Pandemic Response

Climate science was an important, early choice for our keynote topic before the COVID-19 situation heightened, and our Keynote Chair and I felt it was important to keep that focus as the world turned its attention to all things pandemic. Dr. Bjorn Stevens hit exactly the right note – lighthearted, while setting the climate issues we are racing to solve in sharp relief. It’s definitely worth a view or three.

Watch the Keynote

Our More Than HPC Plenary Chairs put together a brilliant panel of experts to discuss how we are using HPC to fight COVID-19 today, and how we might combine it with data science and data collection in the future in such a way that we’re prepared to instantly take action for the next calamity.

Watch the More Than HPC Plenary

A Community, and a Conference, Endures

Delivering SC20 was an incredible challenge for all of us; volunteers, authors, exhibitors, participants, and organizers alike, in a year that devastated us like no other. There will never be enough words to express the appreciation I have for all of you, and for our sponsoring societies, IEEE-CS and ACM, for their continued support.

I hope you’ll take the time to look back upon what we accomplished as a community and realize that congratulations are in order for all of us – to everyone who donated hours, days, weeks, and months of time and effort to prove that anywhere we are, and everywhere we are, we are more than HPC.

In deepest gratitude,
Christine E. Cuicchi
SC20 General Chair

During the SC20 Awards Ceremony, held virtually on Thursday, November 19, 2020, the SC Awards and selected ACM and IEEE-CS awards were presented. Congratulations to all of this year’s awardees!

Society Awards

ACM/IEEE-CS Ken Kennedy Award

Read about the awardee on the ACM website

• Dr. Vivek Sarkar, Georgia Institute of Technology
• Watch the award presentation on YouTube

IEEE-CS Sidney Fernbach Memorial Award

Read about the awardee on the IEEE-CS website

• Dr. Salman Habib, Argonne National Laboratory
• Watch the award presentation on YouTube

ACM Gordon Bell Prize

Pushing the Limit of Molecular Dynamics with Ab Initio Accuracy to 100 Million Atoms with Machine Learning

Weile Jia, University of California, Berkeley; Han Wang, Institute of Applied Physics and Computational Mathematics (Beijing, China); Mohan Chen, Peking University; Denghui Lu, Peking University; Lin Lin, University of California, Berkeley and Lawrence Berkeley National Laboratory; Roberto Car, Princeton University; Weinan E, Princeton University; Linfeng Zhang, Princeton University

ACM Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research

AI-Driven Multiscale Simulations Illuminate Mechanisms of SARS-CoV-2 Spike Dynamics

Rommie Amaro, University of California, San Diego; Arvind Ramanathan, Argonne National Laboratory; Tom Gibbs, NVIDIA; John Stone, University of Illinois at Urbana-Champaign; Jim Phillips, University of Illinois at Urbana-Champaign; Lillian Chong, University of Pittsburgh; Lorenzo Casalino, University of California, San Diego; Abigail Dommer, University of California, San Diego; David Hardy, University of Illinois at Urbana-Champaign; Julio Maia, University of Illinois at Urbana-Champaign; Thorsten Kurth, NVIDIA; Shantenu Jha, Rutgers University

ACM Student Research Competition (Undergraduate)

First Place

Enabling Graph-Based Profiling Analysis Using Hatchet

Ian D. Lumsden, University of Tennessee Knoxville

Second Place

Recovering Silent Data Corruption Through Spatial Prediction

Sarah A. Placke, Clemson University

Third Place

A Framework for Linking Urban Traffic and Vehicle Emissions in Smart Cities

Clark Hathaway, University of Tennessee Knoxville; Sebastian Mobo, University of Tennessee Knoxville

ACM Student Research Competition (Graduate)

First Place

Memory-Centric 3D Image Reconstruction with Hierarchical Communications on Multi-GPU Node Architecture

Mert Hidayetoglu, University of Illinois

Second Place

Communication-Avoiding Large Graph Algorithms for Flow Modeling

Richard Barnes, University of California Berkeley

Third Place

Optimizing Vector Particle-In-Cell (VPIC) for Memory Constrained Systems Using Half-Precision

Nigel P. Tan, University of Tennesse Knoxville

ACM SIGHPC Outstanding Doctoral Dissertation Award

Read about the awardee on the SIGHPC website

• Dr. Patrick Flick, Georgia Institute of Technology

ACM SIGHPC Fellowships in Computational and Data Science

Read about the awardees on the SIGHPC website

• Rachael Aber (PhD, Integrative Biology), Oregon State University
• Clint Cuffy (PhD, Computer Science), Virginia Commonwealth University
• Katherine Cyr (MS, Biostatics), Harvard T.H. Chan School of Public Health
• Emily Elhacham (PhD, Geoscience), Tel Aviv University
• Giulia Guidi (PhD, Computer Science), University of California, Berkeley
• Caitlin Harrigan (PhD, Bioinformatics and Computational Biology), University of Toronto
• Ciara Horne (PhD, Systems Engineering), University of Virginia
• Robert Kravec (MS, Statistics), Duke University
• Kevin McIver (PhD, Mechanical Engineering), Purdue University
• Mira Nencheva (PhD, Psychology), Princeton University
• Ashley Schwartz (PhD, Computational Science), San Diego State University
• Pranjal Vaidya (PhD, Biomedical Engineering), Case Western Reserve University

ACM SIGHPC Certificate of Appreciation

Presented to the authors of the SC19 paper selected for the SC20 Student Cluster Competition Reproducibility Initiative.

MemXCT: Memory-Centric X-Ray CT Reconstruction with Massive Parallelization

Mert Hidayetoglu, University of Illinois; Tekin Bicer, Argonne National Laboratory; Simon Garcia de Gonzalo, University of Illinois; Bin Ren, College of William & Mary; Doga Gursoy, Argonne National Laboratory; Rajkumar Kettimuthu, Argonne National Laboratory; Ian T. Foster, Argonne National Laboratory; Wen-Mei W. Hwu, University of Illinois

ACM/IEEE-CS George Michael Memorial HPC Fellowship

Read about the awardees on the ACM website

• Kazem Chesmi, University of Toronto
• Madhurima Vardhan, Duke University
• Keren Zhou, Rice University

IEEE-CS TCHPC Award for Excellence for Early Career Researchers in High Performance Computing

Read about the awardees on the TCHPC website

• Dr. Catherine Schuman, Oak Ridge National Laboratory
• Dr. Dingwen Tao, Washington State University
• Dr. Pedro Valero-Lara, Oak Ridge National Laboratory

SC Awards

SC Test of Time Award

An Overview of the BlueGene/L Supercomputer (published at SC02)

The BlueGene/L Team (114 authors), accepted by Jose Moreira and Jeffrey S. Vetter

SC Best Paper Award

Petascale XCT: 3D Image Reconstruction with Hierarchical Communications on Multi-GPU Nodes

Mert Hidayetoglu, University of Illinois; Tekin Bicer, Argonne National Laboratory (ANL); Simon Garcia de Gonzalo, Barcelona Supercomputing Center (BSC); Bin Ren, College of William & Mary; Vincent De Andrade, Argonne National Laboratory (ANL); Doga Gursoy, Argonne National Laboratory (ANL); Rajkumar Kettimuthu, Argonne National Laboratory (ANL); Ian T. Foster, Argonne National Laboratory (ANL); Wen-mei W. Hwu, University of Illinois

SC Best Poster Award

Quantum Circuit Optimization with SPIRAL: A First Look

Scott Mionis, Franz Franchetti, Jason Larkin; Carnegie Mellon University

SC Best Visualization Award

The Diurnal Beating of the Amazonian Hydroclimate

David Bock, National Center for Supercomputing Applications (NCSA); Francina Dominguez, University of Illinois; Jorge Eiras-Barca, University of Illinois; Zhao Yang, Pacific Northwest National Laboratory (PNNL)

SC Best Student Paper Award

Scalable yet Rigorous Floating-Point Error Analysis

Arnab Das, University of Utah; Ian Briggs, University of Utah; Ganesh Gopalakrishnan, University of Utah; Sriram Krishnamoorthy, Pacific Northwest National Laboratory (PNNL), Washington State University; Pavel Panchekha, University of Utah

SC Virtual Student Cluster Competition Awards

Highest Linpack Benchmark and Overall Winner

Tsinghua University, China

• Students: Chen Zhang, Jiajie Chen, Yutian Wang, Runxin Zhong, Mingshu Zhai, Zeyu Song
• Advisors: Jidong Zhai, Wentao Han, Lin Gan

Jim Brandt, SC20 Awards Chair

When the dust finally settled, the teams in aggregate used $61.3K of their combined $70.3K budget. Each team had been given a $3,200 initial budget followed by a “surprise” $500 boost 12 hours before the competition’s end. The biggest single-day aggregate spend was the final day of the competition, which saw over half of the total spend being consumed.

As teams worked to complete the various scientific application challenges their spend fluctuated, sometimes swinging wildly up and down. The combined spending rate was $4500/hour at one point. The peak team spending rate was achieved by the Tsinghua University team, who was using $540/hr at one point in the competition.

Throughout the competition all the teams showed poise, collegiality, and a great spirit of friendly competition as they peppered the application experts, and each other, with questions about how to optimize throughput in the cloud via Slack. The Gromacs molecular dynamics application, with a challenge focusing on COVID-19 research, was of particular relevance this year and presented teams with a unique learning experience. Another challenging application was CESM, where the students looked at how to optimize a climate modeling code. The competition was rounded out by the Reproducibility Challenge, where students had to reproduce results from an SC19 paper and the Mystery Application, miniVite, an Exascale Computing Project mini-application.

Although the competition was grueling and much different this year than at past SC conferences, most student teams found participating to be very rewarding. Both advisors and students agreed that everyone learned a lot about how to run HPC workloads in the cloud. Now all that is left for the teams is the announcement of the winner at the SC20 Awards Ceremony on Thursday, November 19 at 3 pm EST.

The VSCC wouldn’t have been possible without the VSCC Committee, the application experts, poster judges, and the VSCC sponsors: Microsoft, Terawe, and ARM. Their support was critical to making the competition a possibility in the cloud.

Scott Michael, SC20 Student Cluster Competition Chair

Verónica G. Melesse Vergara, SC20 Student Cluster Competition Chair

The SC20 Transparency and Reproducibility Initiative undertook a survey in August of this year to study perceptions within the community of the measures implemented by SC to make results from the Technical Program and Workshops more transparent and reproducible. SC’s measures for increased reproducibility take the form of additional information collected at the time of submission of papers to the Technical Program about the experimental activity that supports the results in the manuscript. This information is submitted as an appendix to the manuscript and published as an appendix to the final published version in the SC Conference Proceedings.

The study is timely, as the reproducibility initiative has been in effect since SC15, when authors of SC15 papers were invited to submit an Artifacts Description (AD) Appendix after the conference.

While analysis of the data is ongoing, early findings speak to the strength and value of the sustained effort towards transparency and reproducibility at SC.

After six years of the initiative, a reassuring result of the survey is that a full 90% of the community respondents are aware of the issues around reproducibility. Only 15% think the concerns about reproducibility in science are exaggerated, and only 7% think that the concerns about reproducibility in science do not apply to computer and computational science.

An important impact of the SC reproducibility initiative is whether the reproducibility measures result in published papers having greater impact. The results suggest this to be true. Thirty five percent (35%) of the respondents have used the appendices information from papers in the SC Conference Proceedings in their research.

Food for thought for the future of the SC reproducibility effort: 75% of respondents strongly agreed or somewhat agreed with the statement that transparency (and not reproducibility) is the goal of the SC AD/AE appendices. This result is somewhat surprising and suggests that the visibility that has come through the additional appendix information is overwhelmingly seen as a benefit of the SC reproducibility measures.

The survey was open August 17–31, 2020 and had 204 survey respondents. A manuscript is in preparation that discusses the study, analysis, and detailed conclusions. The study was conducted under Indiana University protocol #2005780098, “Assessing Reproducibility Initiative of IEEE/ACM SC Conference,” led by Beth Plale and Line Pouchard.

Contact the Transparency and Reproducibility Committee for more details.

Beth Plale, SC20 Transparency and Reproducibility Initiative Chair

Line Pouchard, SC20 Transparency and Reproducibility Initiative Evaluation Chair

We look forward to seeing you at SC20 and hosting students from around the world! This blog includes special event highlights and items of interest for students attending the conference. These events and programs are completely free with your registration!

Pro Tips

To make the most of your experience at SC, make sure you have a stable network connection and review the schedule ahead of time to ensure you can make time for sessions you find interesting.

Navigating SC

First time attending SC? No problem. Check out the Navigating SC page to see how SC is organized, and check out the video on the Virtual Event FAQ page to learn more about how to attend virtual SC.

Students@SC Sessions

Visit the SC20 Virtual Platform for the latest details on the Students@SC program.

Students@SC will host sessions via the virtual platform and external meeting platforms during the day. Recordings will be made available soon after each of the live events. Information on how to join all these events will be displayed on the Students@SC page in the virtual platform.

Week 1: November 9–13

Monday, November 9
1 pm ET: Emerging Technologies Panel

Tuesday, November 10
10 am ET: Careers in HPC Panel
1 pm ET: Evolution of a Project

Wednesday, November 11
10 am ET: Thinking in Parallel
1 pm ET: Managing Interactions

Thursday, November 12
10 am ET and 7 pm ET: Perfecting Your Elevator Speech and Growing Your Professional Network Workshop
12:30 pm ET and 9:30 pm ET: Resume Workshop

Week 2: November 17–19

Tuesday, November 17
11 am ET: Funding & Disseminating Research

Wednesday, November 18
11 am ET: Overcoming Struggles & Hurdles in Your Career Pathway

Thursday, November 19
11 am ET: Dealing with Microaggressions Panel

Additional Programming

Virtual Student Cluster Competition

Watch the best and brightest students in HPC build clusters in the cloud in a non-stop competition! More details on how to follow the event will be posted to our Students@SC Page in the virtual conference platform. You can check out the team profiles on the VSCC website, and follow their progress on the live dashboard.

Saturday–Wednesday, November 8–11

Students@SC Alumni Stories

Listen to stories and advice from Students@SC Alumni. There will be a live Q&A session in the second half of each event, so bring your questions about the conference, program, research, and careers!

Tuesday, November 17, 2–3 pm ET
Wednesday, November 18, 3–3:30 pm ET

Student Volunteer Lightning Talks

Listen to fellow students give short, broader talks on a variety of topics from their research. A great way to get to know your peers!

Tuesday, November 17, 10:30–11:00 am ET and 3:30–4:30 pm ET
Wednesday, November 18, 10:45–11 am ET and 3:30–4:30 pm ET
Thursday, November 19, 12:30–1:30 pm ET

Career Building

Job Fair

The SC20 Job Fair is already open and will run through January 15, 2021! SC is inviting attendees of all skill levels (students, post-docs, and professionals) to participate in the Job Fair. As applications and résumés submitted will be provided to recruiting organizations, submit your résumé today and help recruiting organizations find you. Employers will be posting their job listings through January. Search the job postings by skill level, area of interest, and more.

Résumé Workshop

Pre-registration by email is required by November 5.

Have your résumé critiqued by professionals and potential employers! If you plan on participating in the résumé workshop you must submit your résumé to the Job Fair AND email the Students@SC programming chair, Sally Ellingson, at to let her know you want your résumé reviewed at the workshop.

Networking & Mentoring

Guided Interest Groups

Register for a group by November 8.

This year, student programs will be hosting Guided Interest Groups (GIGs). These groups will be small cohorts led be Lead Student Volunteers that will guide you through a variety of topics featured at the SC Conference.

GIG Topics:

• HPC At A Glance: Nodel Level Performance Engineering and Distributed Algorithms, Vijay Thakkar
• The Wonderful World of Machine Learning, Aroua Gharbi
• Storage and I/O for HPC, Jean Luca Bez
• High Performance Computing in Medical Sciences and Health Research, Sadura Akinrinwa

Please take the time to review the different Guided Interest Groups and use the form to submit your selection.

Trivia Party!

November 14, 12–1:30 pm ET and 8–9:30 pm ET

Play some fun, casual trivia with your peers and mentors as a break from all that’s going on at the conference.

Pre-registration is required

Watch Now

For the latest Students@SC Webinars and content for students, visit our YouTube Playlist. Recorded sessions and student interviews will all be published to the official conference YouTube channel.

Social Media

Follow these social media channels dedicated to student content:

• Mentor–Protégé LinkedIn Group
• Students@SC on Facebook
• Students@SC on Instagram
• Students@SC on Twitter

Students@SC Slack

Join our Slack channel to chat with other students participating in a Students@SC program:

Slack Sign-Up

The Test of Time Award (ToTA) recognizes an outstanding paper that has deeply influenced the HPC discipline. It is a mark of historical impact and recognition that the paper has changed HPC trends. We are pleased to announce the selection of the SC02 paper, “An Overview of the Blue Gene/L Supercomputer” as the SC20 ToTA winner.

This work was the first peer-reviewed paper to disclose the Blue Gene/L system. For nearly a decade, the BG series won multiple TOP500 awards and Gordon Bell Prizes (including finalists) and served as a vehicle for many research publications. These papers included topics of node architectures, applications optimization, networks, system administration, fault-tolerance, power management, programming models, file systems, performance tools, and applications science. Additionally, Blue Gene/L was a precursor of the importance of energy-efficiency when it was not a recognized problem in the community, and it is now the dominant constraint for HPC architectures. This paper has had a tremendous and ongoing impact on the design of following supercomputers.

Jose Moreira was the System Software Architect for Blue Gene/L. He is currently a Distinguished Research Staff Member at the IBM Thomas J. Watson Research Center, developing innovations for the POWER line of server processors.

Jeffrey S. Vetter continues to push the forefront of high performance computing as the Section Head for Advanced Computing System Research at Oak Ridge National Laboratory.

Drs. Moreira and Vetter will be presenting the work with several guest speakers joining.

View ToTA in the Schedule

—

Amanda Randles, SC20 Test of Time Award Chair

Recognizing Racism

Recent events, including the horrific and unacceptable killings of Breonna Taylor and George Floyd as well as the protests of police actions that led to their deaths, have heightened our awareness of the ways systemic racism impacts us all. The SC Steering Committee recognizes that systemic and institutionalized racism have been blinding, leading to a failure to create a conference that is inclusive for BIPOC (Black/Indigenous/People of Color) communities. Amidst the global reexamination of systemic and institutional racism, we are addressing these issues within SC’s context to instantiate a thoroughly welcoming and inclusive environment.

Our Pledge

Today, the SC Conference organizers pledge to the SC community to take direct action to improve the racial climate. As a start, we have created a task force on Racial and Ethnic Diversity and Inclusion at SC to help shape and drive our actions moving forward. This task force aims to investigate and offer deliberate actions the conference might pursue while striving for transparency and accountability. Initial recommendations from the task force are already underway. SC20 will publish statistics on the racial and ethnic diversity of conference attendees and the committee is thrilled to provide complimentary technical program registration to students, faculty, and staff from minority-serving institutions.

Future Conferences

As planning continues for future conferences, we are excited to proceed with programs focused on diversity and inclusion such as HPC in the City, HPC Immersion, and the SIGHPC Computing 4 Change Program. We also look forward to continuing to support grassroots organizations that have chosen to host their workshops, Birds of a Feather sessions, and events at the SC conference. Furthermore, the SC Conference organizers are committed to upholding the letter and the spirit of the ACM and IEEE Codes of Conduct which prohibit discrimination, harassment, and bullying against any person for any reason.

The SC Steering Committee and the Task Force are currently developing additional online resources for contacting the committee and providing feedback. These will be posted as soon as the content is finalized.

Those looking to participate in the planning of future SC Conferences can learn more about joining the committee on the Planning Committee Applications web page.

Thank You External Members

The SC Steering Committee would like to acknowledge the external members of the Task Force on Racial and Ethnic Diversity and Inclusion for their time, effort, and dedication in their tremendous work thus far including the crafting of this statement.

• Dorian Arnold, Emory University
• Roscoe Giles, Boston University
• Wendy Wilhelm, Intel Corporation
• William Burke, The George Washington University

Resources

• SC Conference Code of Conduct
• Inclusivity Grants for Complimentary Registration
• Planning Committee Application for Future Conferences

As a week-long event held in different cities every year, the International Conference for High Performance Computing, Networking, Storage, and Analysis, known as SC, provides one of the best opportunities for computer scientists, engineers, network experts, and other HPC researchers from academia, government, and industry to gather and assess the state of the field.

The sheer size and depth of the conference means that a trip to SC likely will require some face time with your manager, to explain how and why attending is worth it for you and for your organization.

Now that SC20 has gone virtual, here’s our top five reasons to attend SC20 that you can use in your appeal to your manager.

1. Full Program

Just as past in-person SC Conferences have been the largest annual gathering of high performance computing, networking, storage, and data analytics professionals, the SC20 Planning Committee has worked tirelessly to bring the broad range of content you’ve come to expect from every SC.

2. Reduced Costs

We’re all disappointed that we aren’t able to come together as a community in the beautiful city of Atlanta, but on the bright side, you can still experience the full breadth of SC’s valuable learning experiences without having to pay for travel and housing, and the registrations fees for SC20 have been greatly reduced to boot.

3. Experience More

While it can be exhilarating running from session to session in a vast convention center, the sheer size of SC and breadth of its content sometimes makes it a challenge to attend everything on your list. As long as you register by December 7, you’ll be able to go back and watch the material you might have missed through the virtual event portal – for six months!

4. Virtual Exhibits

The virtual exhibit hall will be full of the top companies showing off their newest technologies and latest discoveries. There will be opportunities for one-on-one chats, giveaways, on-demand booth sessions, a product showcase, and more. And don’t miss the SCavenger Hunt, available through the mobile app – this will be a fun way to connect with participating exhibitors and win prizes.

5. Share Your Knowledge

State of the Practice Talks will offer material for every level of your business or computing center, from Sysadmins and Education & Outreach professionals to Programmers and Cybersecurity practitioners. Invited Talks will cover cutting-edge and diverse topics including the end (or not) of Moore’s Law, trustworthy modeling, biodiversity, AI, machine learning, and more. The knowledge you bring back from your SC experience can help propel your organization to new levels of greatness.

Register for SC20 Today

—

Charity Plata, SC20 Communications Team Writer

Charity Plata provides comprehensive editorial oversight to Brookhaven National Laboratory’s Computational Science Initiative. Her writing and editing career spans diverse industries, including publishing, architecture, civil engineering, and professional sports. Prior to joining Brookhaven Lab in 2018, she worked at Pacific Northwest National Laboratory primarily within the Advanced Computing, Mathematics and Data Division.

Que’Aire Anderson, a C4C mentor, is studying English and Applied Psychology at the University of California Santa Barbara. When she participated in Advanced Computing for Social Change (ACSC) in 2019, she found that, as a non-STEM major, her interdisciplinary background helped with her understanding of the project, and had a tremendous impact. Yet another in our large group of night owls, Que’Aire likes to search youtube for new recipes. She prefers strategy games, like Mancala — maybe she and Regina can teach the rest of the group how to play. Que’Aire’s role model was (and is) her grandmother, who took care of everyone in the household. We’re glad to have her back for C4C.

Dairian Balai, another C4C mentor, studies Data Science at Chaminade University of Honolulu. Like Que’Aire (above), Dairian is a night owl, and her role model is her grandmother. Her late-night activities include reading or Netflix. Dairian also participated in ACSC last year, and came back because she wants to make sure that others have the same great experience.

Mario Gallegos is a junior at Texas State University, studying Healthcare Administration. During his free time, Mario can be found at the gym or napping. Growing up, his role model was his mother, who overcame all kinds of obstacles and raised three children. We’re sure he will make her proud during C4C.

Cesar Monsalud is a Computer Science and Engineering major at the University of Illinois at Urbana-Champaign. When he’s not studying, or reciting things backwards (like the alphabet), Cesar likes to talk to people — anywhere. Talking to people on the bus, or to the person walking their dog, reminds him that it’s a big world, and we need to recognize the diversity of people’s life experiences. We’re sure he’ll enjoy talking to his C4C teammates about their varied experiences.

Also participating in our virtual C4C this year are Steven Galloway II, Chris Min, Elliott-Zion Mpeye, and Arthur Ymele. We hope they all have a fun and enlightening C4C experience!

Read more Computing4Change student profiles at Part 1 and Part 2 of this series.

—

Melyssa Fratkin, SC20 Communications Chair

 

Steve Conway
Senior Advisor, HPC Market Dynamics, Hyperion Research

An exaflop is an arbitrary milestone, a nice round figure with the kind of symbolic lure the four-minute mile once held. And as we know, there are three stages to these computing milestones that have occurred about once a decade. First comes peak exaflop performance, then a Linpack/TOP500 exaflop, and finally the one that counts most but in the past has been celebrated least: sustained exaflop performance on a full, challenging 64-bit user application.

Not long ago, a fundamental premise underlying advanced supercomputer development was that evolutionary market forces were too slow and governments needed to stimulate revolutionary progress. The idea was that the government would do the heavy lifting to pave the way, and the mainstream HPC market would follow to take advantage of the revolutionary advances. In our annual HPC predictions, Hyperion Research analysts (then at IDC) pointed out the risk that the government-supported high-end HPC market might split off as a separate ecological niche, while the mainstream market continued to evolve on its own inertial path.

That split hasn’t happened. Instead, government officials for the most part have realized that they are no longer the primary drivers of HPC. Market forces have usurped that role. In the worldwide HPC market’s diversification and financial expansion during the past three decades, from about $2 billion in 1990 to more than $27 billion in 2019, government has kept an important role but no longer occupies the kingpin position it once held. Government officials in most HPC-exploiting countries have inflected their strategies to take better advantage of market forces, especially technology commoditization and open standards.

The fact that governments have met HPC market forces partway is a good thing for almost all parties (some legacy codes still need to be modernized). It means that many of the government-supported technology advances for exascale computing will sooner or later benefit the mainstream HPC market, including SMEs that buy only a rack or two of technical servers. Some of these advances may spill over into the larger IT market, well beyond the boundaries of HPC.

That, in turn, means that savvy government officials can help to justify the skyrocketing investments needed for exascale supercomputers by pointing to likely ROI in the larger HPC and IT markets, including potential benefits for industry and commerce.

Measuring these real and potential benefits becomes important. Until recently, even the most powerful, expensive new supercomputers were designed in part as “Linpack machines.” Even though the Top500 Linpack test was never intended to measure the performance of high performance computers on a broad spectrum of workload types, many government funders saw superior Linpack performance as a mark of leadership.

The NCSA “Blue Waters” procurement set a different example by placing overwhelming stress on the assessed needs of user applications, targeting performance gains on these applications as the primary measure of success. Exascale initiatives around the world have largely followed suit, though sometimes with one eye still on the Top500 list. The point here is that, if more buyers of HPC systems at all price points target performance gains on user applications and treat the Linpack benchmark as it was meant to be treated, this should lead to better system balance and wider applicability of HPC systems over time.

First-generation exascale systems have been designed to alleviate existing architectural imbalance, something especially important given the rise of AI and other data-intensive workflows. Studies by Hyperion Research and others confirm that many of the economically important AI applications, including precision medicine, automated driving and Smart City development, will benefit from architectures with sufficient balance to efficiently support concurrent simulation and analytics runs. An important future capability will be to integrate the sometimes orthogonal results of simulation and analytics runs, rather than forcing this to happen inside the brains of researchers.

More good news is that governments around the world have increasingly recognized that HPC is a transformational technology that can boost not only scientific leadership, but also industrial and economic competitiveness. Accompanying this recognition is the notion that HPC is too strategic to outsource to another country, meaning to the U.S. in most cases. Exascale initiatives in Asia and Europe are promoting the development of indigenous technologies, often in conjunction with non-native components. In the short term these indigenous movements may cause some understandable market disruption, but in the end they should increase choices and competition in the global HPC market.

Hyperion Research has said for some years that software advances will be even more important than hardware progress in determining HPC’s future. It’s gratifying to see national and regional exascale initiatives increase funding for exascale software development, although the amounts still seem unequal to the task.

The long-term good news is that HPC has become a mature market, one driven by market forces. That gives strong assurance that the market will behave rationally over time. Demand, in the form of buyer and user requirements, will increasingly win out.