Every time someone sends an email, makes an online bank transaction or accesses information stored across multiple servers, computers must work together behind the scenes.
Making those interconnected systems function reliably is at the heart of distributed computing and it has been a central focus of University of Houston computer scientist Gopal Pandurangan’s research.
Now, Pandurangan, Moores Professor of Computer Science in the College of Natural Sciences and Mathematics, is being recognized for foundational work in the field while launching a new federally funded project aimed at addressing some of distributed computing’s next challenges.
Pandurangan and seven collaborators received the 2026 Edsger W. Dijkstra Prize in Distributed Computing for their paper, “Distributed Verification and Hardness of Distributed Approximation.” The prize recognizes papers whose impact on the theory or practice of distributed computing has stood the test of time.
At the same time, Pandurangan is part of a nearly $1 million National Science Foundation Future CoRe project focused on developing more resource-efficient protocols for distributed systems, with nearly $300,000 supporting the work at UH.
Together, the two accomplishments span more than a decade of research into a simple question: What are the limits of what computers can accomplish when they must work together?
Defining the Limits of Distributed Computing
Pandurangan describes the internet as perhaps the most familiar example of a distributed network. Rather than relying on a single computer, information moves among many interconnected machines that must communicate and coordinate with one another.
“Whenever there are multiple machines involved, you have distributed algorithms coming in,” Pandurangan said.
The research recognized by the Dijkstra Prize dates to work conducted around 2009 and 2010 and published in 2011. Pandurangan and his collaborators developed a methodology for establishing what computer scientists call “lower bounds” — fundamental limits on how efficiently certain problems can be solved in distributed networks.
Knowing those limits is important. If researchers can prove that a problem cannot be solved faster or with fewer resources beyond a certain point, future researchers know where improvement is possible and where it is not.
The techniques introduced in the paper have since become standard tools in distributed computing research and have inspired years of subsequent work.
For Pandurangan, that lasting influence makes the recognition particularly meaningful.
“You don’t do work for the award. You do work because you like it,” he said. “But it really feels good that it’s been recognized.”
Pandurangan accepted the award this summer at the ACM Symposium on Principles of Distributed Computing in London, where he delivered a keynote tracing the history and impact of the research.
Making Distributed Systems More Efficient and Secure
Pandurangan’s newest NSF-funded research turns toward another fundamental challenge: how multiple computers can reliably reach agreement even when some machines fail, behave incorrectly or are compromised.
Known as the Byzantine agreement problem, the challenge is central to the reliability and security of distributed systems.
Consider an online bank transaction. A customer’s account information may be maintained across multiple servers rather than one machine. Those servers need to agree on the transaction and maintain consistent records even if a server fails or communication is delayed. Blockchain systems and other decentralized technologies rely on similar consensus processes.
Through the NSF Future CoRe project, “Resource-Efficient Byzantine Protocols,” Pandurangan and collaborators at the University of New Mexico and Augusta University are working to develop algorithms that can achieve that agreement while requiring fewer resources.
The research will focus on making the protocols more energy efficient, more communication efficient and more secure. One goal is to develop approaches designed to remain secure even against future attacks involving powerful quantum computers.
“At the end of four years, we hope that we will significantly advance the state of the art in this area,” Pandurangan said.
Building the Next Generation of Researchers at UH
The NSF funding will also support the people carrying the research forward.
At UH, the project is expected to support at least one doctoral student and one postdoctoral researcher. One of Pandurangan’s Ph.D. students is already working on problems related to the project.
For Pandurangan, that investment in young researchers is one of the most important outcomes of federal research support.
“It’s not just money,” he said. “You support students, you do postdocs, you actually build a group.”
That connection also brings the story full circle. Research Pandurangan helped develop more than a decade ago has become part of the foundation of his field. The new NSF project allows him and the next generation of computer scientists to build on that foundation — tackling the efficiency, reliability and security challenges facing increasingly interconnected computing systems.