3/10/2026 Jenny Applequist
Each year since 1996, the David J. Kuck Outstanding Ph.D. Thesis Award has been bestowed on one doctoral dissertation in the Siebel School of Computing and Data Science. Jovan Stojkovic, whose research was overseen by Josep Torrellas, has been named the 2025 winner for his work to advance the frontiers of computer architecture for large-scale datacenters. Some of that work is already in use in large-scale datacenters that serve billions of users. In the coming years, Stojkovic plans to broaden his work in this area to include use of AI technologies to speed the discovery of further advances. He is now at Meta and will join the faculty of the University of Texas at Austin later this year.
Written by Jenny Applequist
The 2025 award has gone to work on architecture of large-scale datacenters.
Jovan Stojkovic has been announced as the winner of the Siebel School of Computing and Data Science’s 2025 David J. Kuck Outstanding PhD Thesis Award. The award was established in 1996 in honor of David Kuck, who was a professor of computer science at Illinois from 1965 to 1993. Stojkovic received the award for his doctoral dissertation entitled “Chasing the ‘Tail at Scale’: Toward Cloud-native Architectures.”
As Stojkovic summarized it, the thesis is about “rethinking the design of computer architecture from both hardware and software sides for large-scale datacenters... Taking a look at how we need to design the new generations of servers, of architectures, of processors, so that we can better fit the emerging workloads in datacenters in the cloud.”
More specifically, the research advanced the design of both server hardware and infrastructure software for cloud-native environments to offer good performance, low cost, and strong energy efficiency when emerging workloads such as microservices and serverless computing are being handled. The properties of these workloads are very different from those of traditional applications. For instance, “tail latency” is even more important than average performance, meaning that even the slowest responses must be fast.
Some of the work was done in partnership with industry, specifically with Microsoft, IBM and Intel. Stojkovic did an internship at Microsoft whose results not only contributed to his dissertation, but have been put into production in Microsoft datacenters that serve billions of users around the world.
It was pretty exciting to have an impact... not only through papers and publications and more academic stuff, but also realizing these ideas in real productions, in large-scale datacenters. he said. — Jovan Stojkovic
Stojkovic is currently doing what he calls a “gap year” at Meta. “What we did with Microsoft... and also now when I’m at Meta, I’m trying to take some of the ideas from my thesis and then see how they play within the real world,” he said. “That’s one of the things that I am really proud of: the impact of my work on industry in real-world production.”
In Fall 2026, though, he will return to academia as an assistant professor of computer science at the University of Texas at Austin. He said he’s been thinking hard about what his next research directions will be.
“I want to continue working on building new processors for datacenters and clouds, but I want to expand beyond cloud-native workloads, such as microservices, and also consider AI in the cloud, which is increasingly important,” he said. “How are we going to build infrastructure that can support efficiently all these sorts of very different workloads? And one of the things that I’m pretty excited about, that I started working on and that I want to continue... is how can we leverage AI tools and technologies to design better hardware and software.”
Stojkovic said he was delighted to receive the Kuck Award.
“I was surprised because I know that many fantastic dissertations were out of Illinois this year, so I was happy and excited... that my thesis was given this award,” he said.
Stojkovic was quick to share credit for his success with his PhD advisor, Josep Torrellas, who is the Thomas M. Siebel Chair in Computer Science in the Siebel School of Computing and Data Science.
“When I started my Ph.D., I knew that I wanted to do something related to architecture... but I was for sure not certain which path directly I want to have. And I was not envisioning my dissertation, from the point I started my Ph.D., to be what it was in the end. So it was not a super straightforward road! But I enjoyed it a lot. And it was very exciting... that I was given the opportunity to work on things that excite me; for that, kudos to my advisor! He was very helpful when I needed guidance, but also very open to suggestions, very open to exploring new topics... He was extremely helpful and supportive to make this work and get the best out of it,” he said. “I think we did great work together. It was really fantastic to work with him.”
Torrellas praised Stojkovic highly as well, and noted that his work has already led to multiple patents with IBM and Microsoft. “Jovan provides a blueprint for making cloud-native computing performant, energy-saving and resource-efficient,” he said. “Jovan’s dissertation also opens up new research problems, and provides a stepping-stone for him to lead this research area in the future.”
Grainger Engineering Affiliations
Josep Torrellas is the Thomas M. Siebel Chair in Computer Science in the Siebel School of Computing and Data Science and holds affiliate faculty positions in the Coordinated Science Laboratory and the Department of Electrical and Computer Engineering in The Grainger College of Engineering at the University of Illinois Urbana-Champaign.