【24h】

CUDA usage in electrodynamics and mechatronics

机译:CUDA在电动力学和机电一体化中的使用

获取原文

摘要

IT is a kind of science which develops fast. Nowadays the GPGPU (General Purpose Graphic Processing Unit) based technology is gaining on the importance. Few years ago, GPU was designed only to process the graphic data. The GPGPU enables computation on all kinds of data. This innovation brings computing performance huge boost. The GPGPU have been developed since year 2003. The technology was not available on the market till 2006, when the first GPGPU have been introduced by nVidia (GeForce G80). The architecture (hundreds of cores) enables parallel work together with special programs (the application adapted to GPGPU requirements), indicate huge efficiency. Computer simulations usually absorb considerable amount of computing power and last long. The unfavorable factor can be reduced by implementation of parallel data processing. Part (or the whole) data can be computed on GPGPU instead of CPU. Main aim of the paper is to point the ways of exploiting the potential of GPGPU in daily scientific work. Fundamental research base is analysis of possible usage of GPGPU in electrodynamics and mechatronics. Moreover, investigation is proved by literature research, which shows the profits like time savings and performance boost.
机译:IT是一门发展迅速的科学。如今,基于GPGPU(通用图形处理单元)的技术越来越重要。几年前,GPU仅设计用于处理图形数据。 GPGPU支持对各种数据进行计算。这项创新为计算性能带来了巨大的提升。 GPGPU从2003年开始开发。直到2006年,nVidia推出了第一款GPGPU(GeForce G80),该技术才在市场上推出。该架构(数百个内核)可与特殊程序(适用于GPGPU要求的应用程序)一起并行工作,从而显示出巨大的效率。计算机模拟通常会吸收大量的计算能力,并且会持续很长时间。通过执行并行数据处理可以减少不利因素。可以在GPGPU而不是CPU上计算部分(或全部)数据。本文的主要目的是指出在日常科学工作中开发GPGPU潜力的方法。基础研究基础是分析GPGPU在电动力学和机电一体化中的可能用途。此外,通过文献研究证明了调查结果,它显示出节省时间和提高性能等收益。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号