Tohma Kawasumi
Papers
1
Total Citations
6
H-Index
1
About
Tohma Kawasumi is a leading researcher in heterogeneous computing systems, with a focus on parallelizing compilers and power-reduction techniques for multicore architectures. His most cited work, the OSCAR parallelizing and power-reducing compiler and API, introduced innovative methods for efficiently connecting general-purpose processor cores with diverse accelerators—a critical challenge in modern high-performance computing (HPC), cloud servers, and AI-driven systems like self-driving vehicles and robots. This invited paper, garnering 6 citations, demonstrates his ability to address the growing demand for both high performance and low power consumption in heterogeneous environments. Kawasumi’s contributions are particularly notable for their practical impact on energy-efficient computing, offering a unified software framework that optimizes resource utilization across varied core types. His research bridges the gap between hardware complexity and software usability, making advanced heterogeneous systems more accessible for real-world applications. As the field of heterogeneous computing expands into autonomous systems and AI, Kawasumi’s work on OSCAR continues to influence the design of compilers that balance speed and energy efficiency, solidifying his role as a key innovator in scalable, power-aware computing solutions.
Research Focus
Key Achievements
Top Papers
- 1