by Karen Barrett-Wilt

Where are you coming to Madison from? 
I grew up in the San Francisco Bay Area. For my PhD, I moved out to Carnegie Mellon in Pittsburgh, and most recently I spent a year in California again as a visiting researcher at Google before joining UW–Madison. 

What sparked your interest in computer systems? 
It was a bit of a winding path. I started college intending to study biology, but I ended up doing all the computational work — the algorithms — for a wearable device project in a biology lab, and that pulled me toward computer science instead. As an undergrad, I landed a software engineering internship at Facebook (before it was Meta). I finished my assigned project early and spent the rest of the summer poking around to understand how the company’s systems worked. I remember hearing informally about how their infrastructure recovered from a major outage, and then watching Facebook publish a paper about it months later — that was the moment I realized this was a real field people built careers in. 

I signed up for a required systems class that fall, and thought, wow, this is interesting! I really like knowing how this actually works. 

The professor started mentoring me, bringing me to file-systems conferences and helping me find a summer research position. I’ve been going deeper into systems ever since. 

What do you work on? 
Broadly, I’m a systems researcher focused on sustainable data centers. My work sits at the intersection of distributed, caching, and storage systems, and I like grounding system design in mathematical modeling as well as hardware-software co-design. 

What problem are you trying to solve? 
Data centers run enormous software systems on top of huge amounts of hardware, and historically, systems research has been about using that hardware as efficiently as possible. But two things have changed that: there isn’t enough hardware to keep up with demand, and data centers consume enormous amounts of power. If you look beyond the data center itself to the supply chain, manufacturing and deploying new hardware is a major source of both carbon emissions and hazardous chemical use. A lot of my work tries to understand the whole pipeline and design software systems that change what hardware we actually need, letting us use hardware longer. 

For example, I’m currently working on a paper about giving retired solid-state drives a second life in a different role in the data center rather than discarding them, which is both more sustainable and more cost-effective than always buying something new. 

How does your work connect to the Wisconsin Idea? 
Data centers and the hardware that powers them affect industries and communities well beyond the tech sector, and the sustainability questions I work on — how we build and use hardware more responsibly — have implications far outside a research lab. I’d like my work to have that kind of broader relevance. 

How would you explain that to someone outside computer science? 
Most people use cloud services dozens of times a day without ever thinking about it, like streaming a video or backing up photos from their phone. Those services run on large, sometimes worldwide, distributed systems, and how those systems are designed affects things we all care about: cost, reliability, and increasingly, the environment. My work is about designing those systems so the hardware behind them can be used more efficiently and for longer, instead of being replaced as often as it currently is. 

What do you hope students take away from your classes? 
I want students to see that systems research touches a lot of what happens in their daily lives, often invisibly — the cloud services behind their phone, for instance. And I want them to come away understanding that there are still plenty of open problems and trade-offs in this space left to solve. 

What drew you to UW–Madison? 
When I came out to interview, I already knew several faculty in my specific area, but everyone I met across the department was incredibly welcoming. I also finally had time to explore Madison itself, and I was impressed. One of the things I loved about Pittsburgh during my PhD was that it was green and full of trees, which isn’t something you get in much of California, and Madison has that same quality along with plenty of city life. I do improv comedy, for instance, and there are good places for that here too. 

Had you been to Madison before that? 
My first time on campus was actually for PhD admissions, years before I was hired — a welcome weekend in late winter. It was freezing, and I remember my roommate wanting to go get boba anyway. What struck me was that everyone was still out and about despite the cold; people just accepted it as part of life here. The grad students I met also seemed genuinely happy with their research and the people they worked with, which stuck with me. 

What are you looking forward to in Madison? 
I’m looking forward to continuing improv and getting outdoors — hiking, being outside in general. Madison has a nice mix of outdoorsy things to do and a real city scene, which I’m excited to dig into. 

Outside of research, how do you spend your time? 
I do improv comedy and enjoy getting outdoors.