Supercomputers for Science:
Honoring Tom Schardt’s NASA Legacy
Left: Tom Schardt. Top right: A view of the IBM 3081 Processor Complex in Building 1 at NASA Goddard Space Flight Center, circa 1983 — one of the first systems Tom Schardt worked on in the early 1980s. Bottom right: The Discover supercomputer in NASA Goddard Building 28, circa 2018. NASA images.
The NASA Center for Climate Simulation (NCCS) reflects back on the long career of Tom Schardt, who retired from NASA in December 2020. Tom had an impactful, 37-year career at NASA Goddard Space Flight Center, centered around the support of science through the use of high-end information technology. His career spanned the numerous computer systems deployed at NASA Goddard — from the IBM 3081 processor complex in the early 1980s to the Discover supercomputer in late 2020. Tom’s exceptional technical expertise and leadership, innovation, and mentoring have not only impacted NASA-supported scientific research for the past several decades but also impact the future of high-performance (HPC) computing at NCCS, throughout NASA, and in the expansive research community that NASA supports.
During his four-decade career, Tom significantly contributed to the success of NASA’s missions by ensuring that powerful, resilient computational systems, massive data storage, and responsive support services were available for complex scientific and engineering research. Tom’s technical leadership at NCCS was key to several ongoing, leading-edge science efforts with critical significance. These efforts include very high-resolution Earth System modeling (simulations) run by the NASA Goddard Global Modeling and Assimilation Office (GMAO); thousands of years of climate models run by the NASA Goddard Institute for Space Studies (GISS) used to support the United Nations’ Intergovernmental Panel on Climate Change Assessments (IPCC); ultra high-resolution maps of lunar gravitational fields derived from the Gravity Recovery and Interior Laboratory mission’s observations (GRAIL); and pathfinding artificial intelligence and machine learning (AI/ML) studies exploring light curves from the Transiting Exoplanet Survey Satellite (TESS) that greatly accelerated the identification of binary and multiple-star system candidates.
From left to right: A GMAO image of aerosol optical depth due to black and organic carbon from anomalous wildfire emissions observed at 6:00 GMT on July 21, 2021; a GISS image showing a wide view of six storms on the Earth — Karina, Paulette, Rene, Sally, Teddy, and Vicky — on September 14, 2020 using the brightness temperature of cloud tops using infrared band 13 from the GOES-16 satellite; a gravity map of the Moon using GRAIL observational data; and a two-dimensional projection of the high-dimensional space of TESS light curve representations. NASA images.
Tom’s legacy at NASA and impact on scientific computing is perhaps best understood from the unique perspectives of his many colleagues. By reading their stories, readers can get a fuller measure of Tom’s profound influence inside and well beyond NASA — and the reason why he was recently honored with the prestigious Robert H. Goddard Legacy Award for the NASA Goddard Computational & Information Sciences and Technology Office (CISTO).
The comments below from Tom’s colleagues were gathered from his Robert H. Goddard Legacy Award nomination and from videos recorded at Tom’s retirement event in December 2020.
]]>► CISTO Chief Dan Duffy
Tom’s outstanding ability to work with scientists and translate their requirements into solutions has been critical to NASA throughout his career. He developed close relationships with scientists at GMAO and GISS, two of the largest producers of key Earth science simulation data within NASA. Throughout the years, Tom provided data management services for several science groups, overseeing the exponential growth of data to the current holdings of over 100 petabytes (PBs) of scientific results. To put this in context, if 100 PBs of data was printed out on double-sided paper, the paper stack would be high enough to go back and forth between the Earth and Moon four times!
To support the rapidly growing science requirements, Tom was extremely innovative throughout this career, setting a high standard for others to follow. In the early days of his career, when Tom was in the User Services Group, he became the resident expert on Macintosh architecture when these computers first came out in the 1980s. Through his experience on the Macs, scientific computing began to leverage these systems, and later, Macs became more broadly used throughout science. As Tom’s career progressed, he became involved in large-scale storage, installing the first data management facility (DMF) tape silos at Goddard. These were robotic tape silos that were utilized in operations for over 20 years!
As you can imagine, managing so many diverse and highly complex IT systems — including the Cray Y, J90s, C90s, Sun Microsystems, SGI Altix, HPE, and SGIs — was not always easy. Yet Tom consistently exhibited poise under pressure when it came to working on complex upgrades or troubleshooting problems. The operational systems that Tom supported were used in scientific workflows that required a high level of uptime and availability. If the systems went down, science data did not get generated and distributed to science collaborators. Whenever this occurred, Tom was always quick to address the problems and communicate with the user community.
In 2003, Tom spearheaded a migration from an aging Sun Microsystems computer system to a new Silicon Graphics International (SGI) data migration facility. The new SGI storage tier virtualization software required a significant amount of planning and data migration on Tom’s part. His leadership, management, and technical efforts throughout the migration resulted in minimal impact to the user community while ultimately providing a more capable storage environment.
Tom used his diversity of experience across a wide range of systems to teach and mentor his colleagues. Early in his career, Tom worked with a young co- op student named Bruce Pfaff, who has since become a recognized expert in high-performance computing (HPC). Tom has also mentored and helped develop other civil servants, including Ellen Salmon and Adina Tarshish, now experts in storage and HPC.
In addition to his technical leadership, Tom served on a number of procurement activities. For example, he was asked to provide procurement support for scientific computing by drafting requirements and evaluating proposals on multiple NASA Source Evaluation Boards (SEBs). Tom served on multiple SEB boards in both the HPC and mass storage/peripheral areas for SEWP, one of the largest US government-wide acquisition contracts for IT. In addition, Tom addressed other contractual challenges that would have seriously affected our ability to serve Goddard and NASA scientists. Through Tom’s support, NASA Goddard has maintained an on-site HPC presence for science throughout his entire career.
The legacy of a person can be measured by the impact they have had on others, regardless of whether or not attribution is given to a specific individual. After 37 years of service, Tom retired in December 2020. He exemplifies the legacy of Goddard Space Flight Center: a selfless person who loved to support science without ever asking to be acknowledged.
I want to express my appreciation for Tom’s dedication and support to NASA during his 37 years of service. Personally, Tom has been a friend, mentor, coach, and colleague, and I would not be where I am in my career without Tom. Congratulations Tom, and have fun in your retirement!
– Dan Duffy
► Facilities Operations Manager Steve Majstorovic
A lot of people don’t know this, but I have worked with Tom for almost four decades, from the 1980s to the present [December 2020]. My first recollection of Tom was back in the early 80s at NASA Goddard Space Flight Center’s Building 1, where we were located at that time. He was doing benchmark testing with Vicki Pendergrass on our IBM 3081, and I remember Tom’s boisterous laugh from those days. I could hear it all the way from the other side of the computer room.
I remember the days from Building 1 when you first showed up at Goddard and we would have lunch together with Nancy, you, and a few other people. And I remember you coming to my house to help me build my patio, which is still standing. Hard to believe, right?
The IBM 3081 Processor Complex in Building 1 at NASA Goddard Space Flight Center, circa 1983. NASA photo.
Keep in mind that, back then, there was no internet and there were no cell phones, so the way that people submitted jobs to one of our systems was through a card reader/card deck. And in the early 1970s, remote job entry (RJE) was not invented yet — people had to bring their card decks in. (So, Tom, how old to do you feel now?)
Later, our two [NASA] codes merged and we acquired Building 22 under Milt Halem around 1984, so we were in both Building 1 and Building 22 at that time. In 1989, we moved everything to Building 28, where our NCCS supercomputers are now located.
Now, a lot people think that Tom is a quiet guy. Not so. He can yak with the best of them, and he has a great sense of humor. I was told to keep this short, so Tom, here’s to you! We’re going to miss your laugh.
I hope you enjoy your retirement. It’s been a great half-century, and I hope you have another. Salute!
– Steve Majstorovic
► HPC Lead Laura Carriere
Tom, I wanted to wish you well on your upcoming retirement. We’ll miss your no-nonsense approach to getting to the root of our issues, because everyone needs someone willing to keep the group focused on the cold, hard facts at hand.
You’ve been supportive as I have taken on this role as HPC Lead, answering endless questions. And I’m trying not to take it personally that you chose to retire one year after I started in this role. (It’s you, not me, right?) All kidding aside, I wish you the best as you leave us behind. I fully expect you to gleefully embark on new adventures as long as they don’t require responding to NASA protocols.
– Laura Carriere
► Senior Technical Staff Ellen Salmon
Tom Schardt’s multi-decade career at the NASA Center for Climate Simulation (NCCS) has contributed significantly to the successful efforts of NASA’s science research community by ensuring powerful, resilient computational systems and massive data storage and services. His technical efforts have contributed excellent support in providing system administration and migration efforts as NCCS systems evolved from IBM mainframes and peripherals in the 1980s, to Cray and SGI supercomputers from the 1990s through the 2010s, to the HPE and other high-performance computers and storage resources used at NCCS today, maintaining high levels of system availability, throughput, performance, and responsiveness for NASA’s science researchers.
Tom has been a hands-on system administrator, shepherding the NCCS mass storage system through numerous migrations and upgrades since it was first installed at the NCCS in the 1990s. He has ensured the provision of essential, high-capacity near-online storage of science data in a system that has grown from 500 terabytes (TBs) in 2005 to more than 100 petabytes and 300 million files as of late 2020, routinely handling the transfer of more than 200 TBs of science data daily.
Through the years, Tom has also provided sound guidance and valuable input to numerous source evaluation boards (SEBs); to the NASA Procurement Technical Committee as HPC subject matter expert, streamlining IT acquisition for NASA and other Federal agencies; and to both government and contractor staff by helping shepherd smooth contract transitions.
Tom’s quiet, steadfast efforts, calm presence, exceptional technical expertise, and keen insights have enabled the NASA science community’s fruitful use of NCCS HPC resources to explore challenging research questions and produce trailblazing results for decades. Tom’s huge contributions have also benefitted IT staff within NCCS, at NASA Goddard, and NASA-wide. His determination to help optimize reporting and compliance burdens have both reduced such workloads for IT staff and engendered esteem and thanks from Goddard and Agency compliance staff.
We will miss you, Tom.
– Ellen Salmon
► NASA Advanced Supercomputing (NAS) User Services Director Leigh Ann Tanner
Hey, Tom, so I heard from Dan Duffy that you were retiring, and I thought, “you must be kidding me,” because I think I’ve known you since the middle 1980s, as long as I’ve been at the NASA Advanced Supercomputing Division (NAS) at NASA Ames. Cathy Schulbach from User Services and I would come out and visit quite often at NASA Goddard, and every time we would have questions, it seemed like you always knew the answers to every single thing, whether it was a Goddard problem we were trying to solve, or something from the Agency, or something at NAS. Because as you know, we’ve had a friendly rivalry or competition between NAS and NCCS for the longest time, but you have been so helpful in everything that we’ve tried to do relating to that, helping to move advanced supercomputing forward at both of our NASA facilities. So, all of your knowledge and experience is truly going to be missed.
Hopefully, you will get some travel in your future, maybe not really soon [because of COVID-19 travel restrictions], but maybe later on. I wish you the best!
– Leigh Ann Tanner
► Computer Scientist Adina Tarshish
Tom, I’m going to have to learn how to cope without you somehow. I am not sure how that is going to be. We’re going to miss you terribly.
– Adina Tarshish
► Retired Assistant CISTO Chief Nancy Palm
Steve and I are saying goodbye to you [with audio only] via a computer when we can’t really see ourselves, but here we go. OK, Tom, people don’t know this side of you, so I’m going to divulge. You’re a growly bear on the outside, and a teddy bear on the inside. You’re dependable, and once anyone becomes your friend, you’re their friend forever. You love dogs and babies and nieces and sisters. Tabitha used to love running between your legs when she first started walking. You are such a very unassuming and private person. You avoid all forms of confrontation, except when you get close to someone, then you let us have it. You don’t criticize others, and if I criticize someone, you would say “well, I don’t know about that.”
I was your supervisor, if you remember, from 1992 until 2004, when the branch got dissolved. You used to come into my office, and we would have numerous discussions, debates, arguments, and just some plain conversations. And, during those hated annual reviews, you would come into my office and bulletize a few items, and then I would spend the rest of the hour telling you not to forget about this, that, and the next thing, and you would come back and write me a much better review. When I was sick, Tom, you took care of things every step of the way, including coming to my house to drop off Goddard business mail. Anyway, the fact that you are retiring makes me feel old.
Keep in touch Tom, because I have your number!
– Nancy Palm
► Lead HPC Systems Engineer Bruce Pfaff
Tom and I shared an office when I started at NASA as a contractor and as a co-op. After I graduated, we spent the next 10 years sharing an office, for almost half of my NASA career. As colleagues, Tom and I had many technical discussions and some not-so-technical discussions. Probably the ones I remember most were when Tom, Lyn Gerner and I would sit in the office and debrief after those infamous subcommittee meetings. I really appreciate Tom’s leadership back in those days through all the various struggles we had with technical challenges, through all the computer system upgrades, and everything that we accomplished. I appreciate the example that Tom set for me then and that he continues to set for me and everyone else since then.
Tom, we will all will miss your very direct, no-nonsense perspective on whatever the crisis issue is. I’m pretty sure I can already hear George Rumney asking, “Hmm…how would Tom react?” the next time a challenging new directive comes down from Headquarters.
I will never be able to think about my NASA career without thinking about the part you played in it, and I’m sure I’m not the only one who feels that way. Thanks for being a great example, a great leader, a great friend, and a brother at that time. I wish you all the best in retirement, and God bless you, Tom. Take care.
– Bruce Pfaff
► Senior Systems Engineer Nicko Acks
Tom, I just wanted to thank you for all the years that we’ve worked together. I remember starting at NASA back in 1997 and you were there, working on the J90s. I worked with you on the Cray T3E, the DMF, and all the years with the SGI equipment. Thank you for your guidance and mentoring over the course of my work there. You will definitely be missed: you’re sort of an institution. The NCCS won’t be the same without you. I want to keep in touch.
Hopefully, you have a good retirement and get a chance to relax. Thank you for everything!
– Nicko Acks
► Scientific Data Visualizer Lori Perkins
Tom, I heard that you retired, and I just wanted to take a minute to say thank you to you. Thank you for your kindness and the wonderful way that you treated me over the years. Thank you for answering all of my “stupid questions,” because there were quite a few of those. I would have been lost if you had not been so kind to me and answered them and just really helped me all along the way. I wish you all the best in your retirement. I hope that you get to do fun things, things that maybe you always wanted to do but maybe haven’t had the time. I hope that this new chapter of your life is great. And hey, don’t be a stranger. I’d really like to see what’s going on in your life after retirement.
– Lori Perkins
Sean Keefe, NASA Goddard Space Flight Center