Data: 03 de maio de 2017

Horário: 10h às 12h

Local: Auditório do Prédio CIC/EST – Campus Universitário Darcy Ribeiro – Universidade de Brasília (UnB)

Palestrante: Professor Dr. Koji Nakano - School of Engineering - Hiroshima University (Este endereço de email está sendo protegido de spambots. Você precisa do JavaScript ativado para vê-lo.)

Title:

FPGA vs. GPU: Comparison of Computational Powers

Abstract:

GPUs (Graphics Processing Units) have been used for general
purpose computing these days. Actually, most super computers are
designed using GPUs. Also, FPGAs (Field Programmable Gate Arrays), in
which any circuit can be embedded dynamically by users, have been used
for high performance computing. In this talk, we briefly review
architectures of GPUs and FPGAs, and roughly compare their
computational powers. Also, we introduce several fundamental parallel
computing techniques using GPUs and FPGAs. In particular, we show
FDFM(Few DSP slices and Few Memory blocks) approach for FPGAs and
BPBC（Bitwise Parallel Bulk Computation）technique for GPUs, which the
speaker has developed for high throughput computation.

Short Bio:

Koji Nakano received the BE, ME and Ph.D degrees from Department of Computer Science, Osaka University, Japan in 1987, 1989, and 1992 respectively. In 1992-1995, he was a Research Scientist at Advanced Research Laboratory. Hitachi Ltd. In 1995, he joined Department of Electrical and Computer Engineering, Nagoya Institute of Technology. In 2001, he moved to School of Information Science, Japan Advanced Institute of Science and Technology, where he was an associate professor. He has been a full professor at School of Engineering, Hiroshima University from 2003.  He has published extensively in journals, conference proceedings, and book chapters. He served on the editorial board of journals including IEEE Transactions on Parallel and Distributed Systems, IEICE Transactions on Information and Systems, and International Journal of Foundations on Computer Science. He has also guest-edited several special issues including IEEE TPDS Special issue on Wireless Networks and Mobile Computing, IJFCS special issue on Graph Algorithms and Applications, and IEICE Transactions special issue on Foundations of Computer Science. He has organized conferences and workshops including International Conference on Networking and Computing, International Conference on Parallel and Distributed Computing, Applications and Technologies, IPDPS Workshop on Advances in Parallel and Distributed Computational Models, and ICPP Workshop on Wireless Networks and Mobile Computing. His research interests includes image processing, hardware algorithms, GPU-based computing, FPGA-based reconfigurable computing, parallel computing, algorithms and architectures.