The Paths to HPC series, presented in collaboration with Women in HPC, showcases the women working in high-performance computing. Our hope is that by highlighting these trailblazers—and the sometimes unique paths they followed into the field—other women will feel inspired to envision themselves in similar roles.

Today we talk with Sakshi Mishra, energy optimization and analytics researcher at the US National Renewable Energy Laboratory (NREL).

What was your path to working with HPC?

I grew up in a small town in a developing country (India) where it wasn’t common to encourage girls to take up STEM careers. But I did love math and science and was supported by my family to pursue my interests.

When the time came to choose a major for my bachelor’s degree studies, I was strongly advised against choosing electrical engineering because I was a girl! I was told, “You are not going to be able to work with machines.” Instead, computer science was deemed a better choice. This logic doesn’t make sense to me, to this day!

Though I liked computer science as much as electrical engineering, I decided to major in electrical engineering because they said I couldn’t do it. Fortunately, my parents agreed to support my decision after some initial resistance. I thoroughly enjoyed the courses on distributed energy resources, renewable energy, and power systems.

In this way, my strong inclinations towards the clean energy domain started at the beginning of my bachelor’s education and continues to intrigue and motivate me. Electrical engineering also required me to keep learning to program as part of the curriculum—so I didn’t miss much by not choosing to major in computer science.

My career journey took another turn when I attended Carnegie Mellon University’s (CMU) Energy Science Technology and Policy master’s program. At CMU, I was introduced to the field of artificial intelligence (AI). The idea of blending the science of clean energy with AI to solve climate change problems called to me.

I started training deep learning-based predictive models for energy applications using the GPU on my computer. As the size of the problem grew, my computer was no longer able to support my research activity. It wasn’t until after I started working at NREL that I got access to an HPC system. I quickly became fascinated with the value it has to offer in building AI-based applications for clean energy research and have never looked back.

I was strongly advised against choosing electrical engineering because I was a girl!

Today, I am an energy researcher by profession, power systems engineer by education, data-scientist by curiosity, and licensed electrical engineer by title, and here is my full journey. To sum up my path to HPC: I ended up working with HPC as part of my research that blended AI and clean energy.

What’s cool about working with HPC?

The field of high-performance computing is truly an interdisciplinary one. The skills and knowledge of the subjects I enjoy (math, programming, and computers) are all utilized at the same time. It’s fascinating to take these skills and combine them with energy science.

My current focus is in high-performance technical computing (HPTC), where I build HPC applications for engineering problems, including advanced power systems simulations, as well as training deep learning models for weather and energy forecasting. Another cool reason to work with HPC: since this field is continually evolving, there are always plenty of opportunities for learning new things and growing. It will be pretty hard to reach a plateau anytime soon!

My path to HPC wasn’t a straight line. I stumbled upon it when I discovered the value it has to offer for my clean energy research.

There’s also the continued legacy of the organization I work at. NREL’s supercomputer Eagle is the world’s largest supercomputer dedicated to energy efficiency and renewable energy. The Eagle system’s energy efficiency also sets it apart. Eagle features a warm-water liquid-cooled design that was originally developed for Peregrine (NREL’s older HPC). This concept has now become standard for other data centers. The system is designed to capture 97% of the supercomputer’s waste heat, which can then be made available for use throughout the building in which the system is housed. In this way, the HPC system at NREL seamlessly combines the two areas that I am keenly interested in—HPC and energy efficiency.

For me, probably the coolest thing about working with HPC is the ability to work with cutting edge HPC technology to solve clean energy engineering problems. Having HPC by my side while running complex optimization and simulations makes me way more effective at solving technical conundrums.

In other words, there is a sense of empowerment in having a computational resource like HPC at hand to wrestle with the climate change-related engineering problems I strive to solve as a part of my day-to-day job. It is exciting to see how efficient we can be when human minds team up with the artificial brain of HPC.

Being able to inspire young girls to enter this field increases the meaningfulness associated with my chosen career path.

Another fulfilling aspect of working with HPC is demonstrating that women (although currently a minority) are just as much, if not more, capable of pursuing a career in this field. Being able to inspire young girls to enter this field increases the meaningfulness associated with my chosen career path.

What are some of the challenges you have faced in taking this path?

My path to HPC wasn’t a straight line. I stumbled upon it when I discovered the value it has to offer for my clean energy research. But as a young girl, there was a fair amount of resistance from the community when I embarked on the electrical engineering path.

Another challenge that kept me from starting out with HPC sooner was the resource availability. I didn’t have access to it until I joined NREL. Being an expensive resource, HPC is not as widely accessible as a desktop computer or a laptop. Unless an individual sets out to find a university or organization that offers education and/or careers focused on this path, the probability of a chance encounter into this field is relatively low.

Are there any mentors you would like to thank?

I am thankful to NREL for giving me the opportunity to utilize HPC resources and learn more about this field. I am also grateful to have met and been mentored by wise and kind individuals during my educational journey in this field. Specifically, I would like to thank my research collaborator Praveen Palanisamy who has played a significant role in helping me get oriented in the HPC and AI fields. Also, to my mentors Kate Anderson and Mark O’Malley, who have strongly supported my research at NREL.