期刊文献+

基于VPU加速的虚拟桌面客户端设计

Design of Virtual Desktop Client Based on VPU Acceleration
下载PDF
导出
摘要 在以性价比为导向的基于云的通用VDI(虚拟桌面基础架构)中,客户机往往非常"瘦"。这使得VDI性能低下,只能完成一些最基本的任务。实际应用中,这类VDI经常出现显示卡顿、画面撕裂。在一些复杂的使用环境中,为了给客户端提供更稳定更高质量的显示服务,客户机需要有硬解码能力。论文中设计了搭载高性能VPU(Video Processing Unit,视频处理单元)的客户机,以赋予客户机硬解码特性,这将提高客户机解码性能和减小CPU负荷,从而改善客户端显示质量。另外,使用了改进的虚拟化协议和经过优化的VPU驱动程序,减少了服务器负载和网络带宽的消耗。实验结果表明,与没有获得VPU加速的客户机相比,搭载VPU的客户机显示帧数提升高达400%,画面也更稳定流畅。基于此,又可以进一步提高视频流数据的压缩比,从而使网络带宽消耗减少高达44%。 In a cloud-based general-purpose VDI(virtual desktop infrastructure)that is cost-effective,clients are often very "thin."This makes VDI performance low and can only complete some of the most basic tasks. In practical applications,such VDIs often appear to be stuck and tearing. In some complex environments,the client needs hard decoding capability in order to provide the client with a more stable and higher quality display service. In this paper,a client equipped with a high-performance VPU(Video Processing Unit)is designed to give the client a hard decoding feature,which will improve the client decoding performance andreduce the CPU load,thereby improving the client display quality. In addition,an improved virtualization protocol and optimized VPU driver are used,which will reduce the server load and network bandwidth consumption. The experimental results show that compared with the client that does not get VPU acceleration,the number of display frames of the VPU-equipped client is increasedby up to 400% and the display is more stable and smooth. Based on this,the compression ratio of video stream data can be further improved,thereby reducing network bandwidth consumption by as much as 44%.
作者 余启家 熊庭刚 张宇 YU Qijia;XIONG Tinggang;ZHANG Yu(Wuhan Digital Engineering Institute,Wuhan 43007)
出处 《舰船电子工程》 2018年第8期12-16,共5页 Ship Electronic Engineering
关键词 云计算 虚拟桌面基础架构 视频处理单元 cloudcomputing VDI VPU
  • 相关文献

参考文献2

二级参考文献31

  • 1VARIA J. Cloud architectures - Amazon Web services [ EB/OL]. [ 2009 - 03 - 01 ]. http://acmbangalore, org/events/monthly-talk/ may-2008 --cloud-architectures---amazon-web-services. html. 被引量:1
  • 2BRYANT R E. Data-intensive supercomputing: The case for DISC, CMU-CS-07-128 [ R]. Pittsburgh, PA, USA: Carnegie Mellon University, Department of Computer Science, 2007. 被引量:1
  • 3SZALAY A S, KUNSZT P, THAKAR A, et al. Designing and mining multi-terabyte astronomy archives: The sloan digital sky survey [ C]//Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data. New York: ACM Press, 2000:451 - 462. 被引量:1
  • 4BARROSO L A, DEAN J, HOLZLE U. Web search for a planet: The Google cluster architecture [ J]. IEEE Micro, 2003, 23(2) : 22 -28. 被引量:1
  • 5GILES J. Google tops translation ranking [ EB/OL]. (2006 - 11 - 06) [ 2009 - 03 - 06 ]. http://www, nature, com/news/2006/ 061106/full/news061106-6. html. 被引量:1
  • 6维基百科.Cloud computing [ EB/OL]. [ 2009 - 03 - 10]. http://en. wikipedia, org/wiki/Cloud_computing. 被引量:1
  • 7中国云计算网.什么是云计算?[EB/OL].(2008-05-14)[2009-02-27].http://www.cloudcomputing-china.cn/Article/ShowArticle.asp?ArticleID=1. 被引量:18
  • 8VAQUERO L M, RODERO-MERINO L, CACERES J, et al. A break in the clouds: Towards a cloud definition [ J]. ACM SIGCOMM Computer Communication Review, 2009, 39(1): 50-55. 被引量:1
  • 9WEISS A. Computing in the clouds [ J]. ACM Networker, 2007, 11(4): 16 -25. 被引量:1
  • 10GRIFFITHS A, METHERALL G. Cluster intereonnection networks [ EB/OL]. (2000 -09 -01)[2009 -03 -03]. http://www, gridbus. org/-raj/csc433/ClusterNets, pdf. 被引量:1

共引文献930

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部