NFV数据平面的网络性能优化技术

黄昱恺; 耿金坤; 令瑞林; 李丹

doi:10.11959/j.issn.1000-0801.2017088

您当前的位置：

首页 >

文章列表页 >

NFV数据平面的网络性能优化技术

专题：SDN/NFV | 更新时间：2024-06-05

- NFV数据平面的网络性能优化技术
- Performance optimization solutions for NFV data plane
- 电信科学 2017年33卷第4期页码：65-70
- 作者机构：
- 作者简介：
  
  [ "黄昱恺（1992-），男，清华大学硕士生，主要研究方向为软件定义网络、网络功能虚拟化、高性能网络协议栈设计与实现。" ]
  [ "耿金坤（1994-），男，清华大学硕士生，主要研究方向为高性能网络协议栈的设计与实现、数据中心带宽资源分配方案的设计与实现。" ]
  [ "令瑞林（1992-），男，清华大学硕士生，主要研究方向为数据中心网络、网络安全与网络性能优化技术等。" ]
  [ "李丹（1981-），男，清华大学计算机系特别研究员、博士生导师，主要从事网络体系结构、网络协议与网络系统的研究工作。担任国家“973”计划（青年科学家专题）项目“软件定义的云数据中心网络基础理论与关键技术”首席科学家，获得2014年清华大学“学术新人奖”，获得2015年国家优秀青年科学基金资助。主持或参加国家科研项目10多项，曾担任国际学术期刊IEEE Transactions on Computer编委（2013-2015年）。《电信科学》杂志编委。发表论文50多篇，论文被引用2 000多次，获得国内外专利20多项。" ]
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2017088
  中图分类号： TP393
- 网络出版日期：2017-04，
  
  纸质出版日期：2017-04-20
- 稿件说明：
移动端阅览
黄昱恺, 耿金坤, 令瑞林, 等. NFV数据平面的网络性能优化技术[J]. 电信科学, 2017,33(4):65-70.

Yukai HUANG, Jinkun GENG, Ruilin LING, et al. Performance optimization solutions for NFV data plane[J]. Telecommunications science, 2017, 33(4): 65-70.
黄昱恺, 耿金坤, 令瑞林, 等. NFV数据平面的网络性能优化技术[J]. 电信科学, 2017,33(4):65-70. DOI： 10.11959/j.issn.1000-0801.2017088.

Yukai HUANG, Jinkun GENG, Ruilin LING, et al. Performance optimization solutions for NFV data plane[J]. Telecommunications science, 2017, 33(4): 65-70. DOI： 10.11959/j.issn.1000-0801.2017088.

摘要

近年来，NFV因为其部署灵活、成本低廉等优良特性，受到了学术界与工业界的广泛关注。然而，性能问题在NFV发展过程中变得愈加明显，逐渐成为制约NFV技术进步的主要瓶颈。针对影响NFV数据平面网络性能的主要因素进行了归纳概述，分析了NFV的性能瓶颈所在，并在此基础上进一步归纳了相应的优化方法。最后，简要介绍了当前比较成熟的两种系统架构，能够显著提升NFV数据平面的网络性能。

Abstract

In recent years

network function virtualization (NFV) is arousing concern widely from both academia and industry due to its flexible deployment and low cost.However

the performance bottleneck becomes increasingly distinctive and hinders the progress of NFV technology.The major factors were studied which could seriously affect the performance of NFV data plane.Firstly

an analysis was conducted on the causes for various kinds of performance overheads.Then

the optimization solutions were summarized to mitigate these overheads.Finally

a brief introduction was made on two types of architecture

which could significantly improve the performance of NFV data plane.

关键词

Keywords

references

李晨，段晓东，陈炜，等 . SDN和NFV的思考与实践 [J ] . 电信科学， 2014 , 30 ( 8 ): 23 - 27 .

LI C , DUAN X D , CHEN W , et al . Thoughts and practices about SDN and NFV [J ] . Telecommunications Science , 2014 , 30 ( 8 ): 23 - 27 .

赵慧玲，史凡 . SDN/NFV的发展与挑战 [J ] . 电信科学， 2014 , 30 ( 8 ): 13 - 18 .

ZHAO H L , SHI F . Development and challenge of SDN/NFV [J ] . Telecommunications Science , 2014 , 30 ( 8 ): 13 - 18 .

HWANG J , RAMAKRISHNAN K K , WOOD T . NetVM:high performance and flexible networking using virtualization on commodity platforms [J ] . IEEE Transactions on Network and Service Management , 2015 , 12 ( 1 ): 34 - 47 .

FIGURE E . Eliminating receive livelock in an interrupt-driven kernel [J ] . ACM Transactions on Computer Systems , 1997 , 15 ( 3 ): 217 - 252 .

YANG J , MINTURN D B , HADY F . When poll is better than interrupt [C ] // FAST , February 14 - 17 , 2012 , San Jose,CA,USA . New York : ACM Press , 2012 : 3 .

WU W , CRAWFORD M , BOWDEN M . The performance analysis of Linux networking-packet receiving [J ] . Computer Communications , 2007 , 30 ( 5 ): 1044 - 1057 .

KOH Y , PU C , BHATIA S , et al . Efficient packet processing in user-level Oses:a study of UML [C ] // The 31th IEEE Conference on Local Computer Networks , November 14 - 16 , 2006 , Sydney,Australia . New Jersey : IEEE Press , 2006 : 63 - 70 .

DEAN J , . Designs,lessons and advice from building large distributed systems [J ] . Keynote from LADIS , 2009 ( 1 ).

DINIZ P C , RINARD M C . Lock coarsening:eliminating lock overhead in automatically parallelized object-based programs [J ] . Journal of Parallel and Distributed Computing , 1998 , 49 ( 2 ): 218 - 244 .

BOYD-WICKIZER S , CLEMENTS A T , MAO Y , et al . An analysis of Linux scalability to many cores [C ] // Usenix Symposium on Operating Systems Design ＆ Implementation , October 4 - 6 , 2010 , Vancouver,BC,Canada . New York : ACM Press , 2010 : 86 - 93 .

KAMRUZZAMAN M , SWANSON S , TULLSEN D M . Inter-core prefetching for multicore processors using migrating helper threads [J ] . ACM SIGPLAN Notices , 2011 , 46 ( 3 ): 393 - 404 .

KAZEMPOUR V , KAMALI A , FEDOROVA A . AASH:an asymmetry-aware scheduler for hypervisors [J ] . ACM SIGPLAN Notices , 2010 , 45 ( 7 ): 85 - 96 .

VAHDAT A , YOCUM K , WALSH K , et al . Scalability and accuracy in a large-scale network emulator [J ] . ACM SIGOPS Operating Systems Review , 2002 , 36 ( SI ): 271 - 284 .

EMMERICH P , RAUMER D , WOHLFART F , et al . erformance characteristics of virtual switching [C ] // 2014 IEEE 3rd International Conference on Cloud Networking (CloudNet) , October 8 - 10 , 2014 , Luxembourg . New Jersey : IEEE Press , 2014 : 120 - 125 .

LIN X , CHEN Y , LI X , et al . Scalable kernel TCP design and implementation for short-lived connections [C ] // International Conference on Architectural Support for Programming Languages ＆ Operating Systems , April 2 - 6 , 2016 , Atlanta,Georgia,USA . New York : ACM Press , 2016 : 339 - 352 .

LEVINTHAL D . Performance analysis guide for intel core i7 processor and intel XEON 5500 processors [J ] . Intel Performance Analysis Guide , 2009 ( 30 ): 18 .

浏览量

831

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

车联网新型基础设施跨域协同部署研究

平台化服务网络——新型移动通信系统架构研究

全国携号转网服务方案设计与政策研究

全息通信对未来网络的需求与挑战

基于NFV的边缘计算承载思路