Header Image

Kentaro Sano

team leader at R-CCS, Riken

Processor Research Team - Center for Computational Science, Riken

Dr. Kentaro Sano received his Ph.D. from GSIS, Tohoku University, in 2000. Since 2000 until 2005, he had been a Research Associate at Tohoku University. Since 2005 until 2018, he has been an Associate Professor at Tohoku University. He was a visiting researcher at the Department of Computing, Imperial College, London, and Maxeler corporation in 2006 and 2007. Since 2017 until present, he has been a team leader of a processor research team at R-CCS, Riken. His research interests include FPGA-based high-performance reconfigurable computing systems especially for scientific numerical simulations and machine learning, high-level synthesis compilers and tools for reconfigurable custom computing machines, and system architectures for next-generation supercomputing based on the data-flow computing model.

Remote talk: FPGA-based Data-Flow Computing for High-Performance Numerical Simulation

Data-flow computing with FPGAs (Field-Programmable Gate Arrays) is promising to further scale computational performance after performance improvement by multi-core scaling becomes difficult. In our team, we have developed a system with FPGAs and a data-flow compiler which generates a pipelined custom hardware module to be embedded onto an FPGA and executed as stream computing. In this talk, I introduce the system, the data-flow compiler, and the case study of FPGA-based Tsunami Simulation, following recent trend of FPGA’s application for high-performance computing.