Gina Tourassi

Congratulations on your selection as a 2024 HPCwire Person to Watch and, of course, being named an IEEE Fellow in January and becoming associate laboratory director for the Computing and Computational Sciences Directorate (CCSD) at Oak Ridge National Laboratory in December. Let’s start with your new role at CCSD — can you please briefly describe it and share your early thoughts on near-term goals for CCSD?

As the associate laboratory director for computing and computational sciences at ORNL, I am responsible for overseeing efforts in key areas aligned with the laboratory’s and the Department of Energy’s computing missions, including high-performance computing, artificial intelligence, quantum information science, and their impactful translation for scientific discovery and technical advances.

In the first three months in my new role, I prioritized engaging with experts across the directorate to deeply understand our strengths and opportunities for advancement. I’m proud of our team that recently deployed Frontier — the world’s first exascale system dedicated to open science — and they immediately got to work on developing the laboratory’s next supercomputer, among other long-term projects.

Meanwhile, our computational scientists tackle high-impact societal problems, from investigating the inner mechanisms of complex diseases to understanding climate change or the behavior of supernovae and other cosmic phenomena. Deploying and operating unique computing resources and pursuing bold questions requires cultivating scientific excellence. Thus, we remain committed to recruitment and development programs that empower staff to thrive individually and together at every career stage.

Looking at the effort to stand up Frontier, the first U.S exascale system, what surprised you most and what went exactly according to plan? Can you provide a sense of a range of the science problems now being tackled on Frontier, and maybe some insight into its workload; do most jobs take a large fraction of Frontier’s resources, or are jobs of varying sizes also run at the same?

Bringing Frontier online was truly a herculean effort for ORNL and our vendor partners. At least 70% of the project happened during the COVID-19 pandemic, which presented significant challenges, from supply chain delays to workforce health and retention concerns, logistics complications, and procurement issues from market shifts. But despite it all, we were able to deliver Frontier on time, within budget, and within scope.

Frontier is available to users to solve computationally intensive scientific problems and to accelerate technology innovation for industry partners. Half of our users are academic scientists, while the rest are from national labs, other federal agencies, and industry. OLCF users are advancing a wide range of science domains, including energy production and environment exploration, materials science, biology, and fundamental explorations of science from nuclear matter to dark energy in the universe.

Frontier also supports privacy-preserving computing, enabling computational campaigns with sensitive or proprietary data. A team from GE Aerospace is currently using Frontier to help develop a revolutionary new open fan engine architecture. The team reported the new engine design could help reduce carbon dioxide emissions by more than 20%.

The OLCF aggressively prioritizes capability jobs, or any single job that uses at least 20% of the leadership system’s available nodes, in our scheduling system. More than 70% of our jobs on Frontier in 2023 represented capability usage.

It’s hard to believe that planning for Frontier’s follow-on system — OLCF-6 — is already underway, with technical requirements having been released for comment in January. The RFP is set to go out later this year and the new supercomputer is scheduled for delivery in the 2027 timeframe. What can you tell us about OLCF-6 goals? Broadly, how will it differ from Frontier, and are there key technologies that have been targeted?

We are in the process of developing the next system. The Request for Proposal (RFP) release is scheduled for May 2024. One key feature of OLCF-6 includes integration of advanced workflows that facilitate tight coupling of experimental and observational facilities with exascale computing resources to further accelerate discovery in an Integrated Research Infrastructure (IRI) ecosystem. IRI will allow for new classes of applications to take advantage of leadership capabilities for extreme scale end-to-end scientific campaigns.

Additionally, the OLCF has been a leader in energy efficient computing, and will continue to drive U.S. leadership in energy efficiency holistically across infrastructure, hardware, software, algorithms, and applications. This is an important driver for OLCF as we are seeing rapid growth of computational campaigns that require tight coupling of extreme scale modeling and simulation with large-scale AI. OLCF-6 will enable the U.S. to lead next generation technology for interoperable and disaggregated computing hardware, extreme bandwidth and interconnects, and integration of accelerators (including AI and quantum devices). OLCF-6 aligns with DOE and national needs, enabling computational science campaigns that cannot be carried out anywhere else in the world while advancing U.S. competitiveness and security.

You’ve said that continuing ORNL’s leadership in AI and quantum technology research are priorities. Could you expand on that and cite example areas in each discipline that will flourish at ORNL?

ORNL’s AI Initiative encompasses a broad range of efforts all aimed at accelerating the security, trustworthiness, and energy-efficiency of AI and its applications. The initiative leverages the lab’s expertise, computing assets, and facilities to accelerate innovations and strengthen national security. Two pillars underpin this effort: real-world applications and cross-disciplinary collaboration.

The application focus deploys AI in three high-impact areas: speeding up scientific discovery, augmenting experimental facilities, and advancing national security. By targeting concrete use cases, we can catalyze progress. In parallel, we are enabling knowledge-sharing across fields to spark creative breakthroughs. Last year, the laboratory also launched the Center for AI Security Research, or CAISER, to expand the AI Initiative’s purview to encompass research related to ensuring individual privacy and global security, as well as preventing other potential vulnerabilities and threats that could stem from the widespread adoption of AI technologies. With interdisciplinary thinking and targeted applications advancing in tandem, ORNL’s AI initiative aims to fully harness AI’s potential for transformative change.

On the quantum side, ORNL is currently developing a quantum roadmap to guide the laboratory’s strategies in driving impactful integration of the lab’s research efforts in quantum materials, networks, sensors, and computing. Key hubs of quantum innovation at the laboratory include the Quantum Science Center, a Department of Energy Quantum Information Science Research Center that brings together researchers from national laboratories, universities, and industry partners, and the Quantum Computing User Program, which provides researchers with access to quantum computing resources.

ORNL researchers are already testing quantum technologies on a real-world scale in Chattanooga, Tennessee, via a strategic partnership with the Electric Power Board (EPB) of Chattanooga, which hosts the nation’s only utility quantum network. Building on 10 years of joint energy-related research, EPB and ORNL recently launched a new Collaborative for Energy Resilience and Quantum Science initiative. The initiative will focus on utilizing Chattanooga’s highly advanced and integrated energy and communications infrastructure to develop technologies and best practices for enhancing the resilience and security of the national power grid while accelerating the commercialization of quantum technologies.

What inspired you to pursue a career in STEM, and what advice would you give to young people who wish to follow in your footsteps?

As I progressed through school, I naturally gravitated towards math, chemistry, and physics courses. My methodical way of thinking coupled with enjoyment of analytical problem-solving led me to a career in STEM. I was fortunate to have a supportive family that encouraged me to dream big and take risks. My advice to those interested in STEM is to follow what captivates you most. Allow your innate curiosity about the world to drive your path, regardless of any barriers that feel discouraging. Stay resilient — failure is part of the learning process. Progress depends on challenging the status quo. Life is full of possibilities!

Outside of the professional sphere, what can you tell us about yourself – unique hobbies, favorite places, etc.? Is there anything about you your colleagues might be surprised to learn?

I don’t have any wild or unique hobbies. When I’m not immersed in my professional work, you can usually find me in the kitchen trying new recipes. My favorite part of cooking is how it brings people together to laugh, connect, and indulge. The food almost feels secondary to the magical atmosphere it cultivates. Whether we’re gathered around the table at my place or at a seaside restaurant overlooking the Mediterranean on vacation, that quality time shared with loved ones over delicious fare is what really nourishes me outside of work. As for surprises, I tend to be a ‘what you see is what you get’ kind of person.