arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.04174cs.DC
收藏

应用程序功能块的自动GPU和FPGA卸载的评估

Evaluation of Automatic GPU and FPGA Offloading for Function Blocks of Applications

Yoji Yamato

摘要:

近年来,使用FPGA、GPU的系统由于其与CPU相比的功率效率等优点而增加。然而,在FPGA和GPU等系统中使用需要了解硬件特定的技术规范,如HDL和CUDA,这是一个很高的障碍。基于这种背景,我以前提出了环境自适应软件,它可以根据要放置的硬件自动转换,配置和高性能操作一次编写的代码。作为这个概念的一个元素,我提出了一种方法来自动卸载CPU的应用程序源代码的循环语句到FPGA和GPU。在本文中,我提出并评估了一种卸载功能块的方法,这是一个更大的单元,而不是应用程序中的单个循环语句,通过自动卸载到GPU和FPGA来实现更高的速度。我实现了所提出的方法,并与现有的应用程序卸载到GPU进行评估。

英文摘要:

In the recent years, systems using FPGAs, GPUs have increased due to their advantages such as power efficiency compared to CPUs. However, use in systems such as FPGAs and GPUs requires understanding hardware-specific technical specifications such as HDL and CUDA, which is a high hurdle. Based on this background, I previously proposed environment adaptive software that enables automatic conversion, configuration, and high-performance operation of once written code according to the hardware to be placed. As an element of the concept, I proposed a method to automatically offload loop statements of application source code for CPU to FPGA and GPU. In this paper, I propose and evaluate a method for offloading a function block, which is a larger unit, instead of individual loop statements in an application, to achieve higher speed by automatic offloading to GPU and FPGA. I implement the proposed method and evaluate with existing applications offloading to GPU.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑