2020/03/09 by Yoji Yamato, Yamato, Yoji
Computer Science · Engineering · #Digital Transformation in Industry #Distributed #Embedded Systems Design Techniques #FOS: Computer and information sciences #IoT and Edge/Fog Computing #Manufacturing Process and Optimization #Parallel #Parallel Computing and Optimization Techniques #Software Testing and Debugging Techniques #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.2005.04174
openalex publication_date 2020/03/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In the recent years, systems using FPGAs, GPUs have increased due to their\nadvantages such as power efficiency compared to CPUs. However, use in systems\nsuch as FPGAs and GPUs requires understanding hardware-specific technical\nspecifications such as HDL and CUDA, which is a high hurdle. Based on this\nbackground, I previously proposed environment adaptive software that enables\nautomatic conversion, configuration, and high-performance operation of once\nwritten code according to the hardware to be placed. As an element of the\nconcept, I proposed a method to automatically offload loop statements of\napplication source code for CPU to FPGA and GPU. In this paper, I propose and\nevaluate a method for offloading a function block, which is a larger unit,\ninstead of individual loop statements in an application, to achieve higher\nspeed by automatic offloading to GPU and FPGA. I implement the proposed method\nand evaluate with existing applications offloading to GPU.\n