Poster 86: High-Performance Custom Computing with FPGA Cluster as an Off-Loading Engine

Authors: Takaaki Miyajima (RIKEN Center for Computational Science (R-CCS)), Tomohiro Ueno (RIKEN Center for Computational Science (R-CCS)), Jens Huthmann (RIKEN Center for Computational Science (R-CCS)), Atsushi Koshiba (RIKEN Center for Computational Science (R-CCS)), Kentaro Sano (RIKEN Center for Computational Science (R-CCS)), Mitsuhisa Sato (RIKEN Center for Computational Science (R-CCS))

Abstract: A heterogeneous system with Field Programmable Gate Arrays (FPGAs) is gathering attention in High-Performance Computing (HPC) area. We have been researching and developing a high-performance custom computing with FPGA cluster as an off-loading engine for an existing HPC system. In this research poster, we describe the motivation of our research and present research topics on a software bridge between the FPGA cluster and existing HPC servers, and dedicated inter-FPGA networks.

Best Poster Finalist (BP): no

Poster: PDF
Poster summary: PDF


Back to Poster Archive Listing