Uzi Vishkin, a professor of electrical and computer engineering with an appointment in the University of Maryland Institute for Advanced Computer Studies, has received the 2025 Association for Computing Machinery Symposium on Parallelism in Algorithms and Architectures (SPAA) Parallel Computing Award.
Presented annually, the award recognizes individuals for a singular breakthrough or a career of influential contributions to the SPAA community.
Vishkin is widely regarded as a pioneer in parallel computing. He helped establish the parallel random-access machine (PRAM) algorithmic theory—a foundational model that simplifies the design and analysis of parallel algorithms. He was recognized at the symposium for his formative work-depth think-parallel framework, many fundamental parallel algorithms, the PRAM-on-chip system framework, and its commitment to silicon.
“This is a very special type of recognition,” Vishkin says. “Many of the papers used the language and concepts I introduced, which is deeply satisfying.”
He adds that the award felt especially meaningful because it came from a community he helped shape from its earliest days.
Vishkin delivered a keynote address at the SPAA conference on July 29 in Portland, Oregon, tracing the early foundations of parallel computing and emphasizing the need for future computer systems to be designed with flexibility and “serendipity” in mind—so they can support applications that don’t yet exist, particularly in the era of AI.
“Part of the interesting thing was to share with the younger audience how the things they take for granted today emerged—and why they maybe shouldn’t be taken for granted,” Vishkin says.
Michael Bender, the John L. Hennessy Chaired Professor of Computer Science at Stony Brook University and SPAA Steering Committee Chair, introduced Vishkin, calling the choice to honor him “one of the easiest decisions to make.”
“Uzi is one of the pioneers of parallel algorithms research,” Bender says. “His seminal contributions played a leading role in shaping what parallel computing means today. Essentially everyone in the conference uses ideas that were developed by Uzi.”
Bender also praised Vishkin’s commitment to important scientific problems “even when it wasn’t popular,” something he tries to emulate in his own research.
Vishkin’s research in parallel computing and parallel computer architecture has earned international recognition. In 1997, he and his team introduced the XMT (Explicit Multi-Threading) desktop supercomputer concept—later known as PRAM-On-Chip—which took a novel approach by using the parallel algorithmic theory to inform both hardware and software design.
By 2007, the XMT prototype featured 64 parallel processors and enabled significantly more accessible and cost-effective programming for software developers.
Vishkin’s 2005 inventions for integrating parallel processing accelerators into the central processing unit (CPU) helped usher in a new area of computer design. The best-known outcome—CPUs coupled with integrated GPUs (iGPUs) has been implemented in more than a billion devices, from desktops to laptops, since the 2010s. Last year, Vishkin was named a Fellow of the National Academy of Inventors.
—Story by Maria Herd and Zsana Hoskins, UMIACS communications group