首页 | 本学科首页   官方微博 | 高级检索  
     

通用平台高性能可扩展网络地址转换系统
引用本文:李峻峰,李丹,黄昱恺,程阳,令瑞林.通用平台高性能可扩展网络地址转换系统[J].北京邮电大学学报,2021,44(2):14-19.
作者姓名:李峻峰  李丹  黄昱恺  程阳  令瑞林
作者单位:清华大学 计算机科学与技术系, 北京 100084
基金项目:国家重点研发计划项目(2018YFB1800100);广东省重点领域研发计划项目(2018B010113001);国家自然科学基金项目(61432002,61772305,61672499)
摘    要:为了提高通用平台网络地址转换(NAT)的性能,设计了高性能可扩展网络地址转换系统——Quick NAT.高速网络地址转换查表算法将原始规则表划分为若干子表,并采用哈希查表算法极大地提高了NAT规则的查找效率;为了充分发挥多核中央处理器(CPU)和多队列网卡的性能优势,设计了高效的并行架构,使用本地化的连接记录表和基于比较并交换原子操作的无锁NAT规则表,避免了多CPU核访问修改全局表带来的锁开销;借助轮询取代中断、越过内核等机制,全程使用指针操作数据包,实现零拷贝,进一步降低开销.实验结果表明,Quick NAT可以极大地提高NAT查表的效率和吞吐量,具有较强的多核可扩展性,能够在10 Gbit/s的网络环境下实现64 byte小包线速.

关 键 词:网络地址转换  性能优化  哈希算法  数据结构本地化  
收稿时间:2020-08-20

High Performance and Scalable NAT System on Commodity Platforms
LI Jun-feng,LI Dan,HUANG Yu-kai,CHENG Yang,LING Rui-lin.High Performance and Scalable NAT System on Commodity Platforms[J].Journal of Beijing University of Posts and Telecommunications,2021,44(2):14-19.
Authors:LI Jun-feng  LI Dan  HUANG Yu-kai  CHENG Yang  LING Rui-lin
Affiliation:Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
Abstract:Quick network address translation (NAT) is proposed to improve the network performance of the NAT system on the commodity server by three ways. Firstly, the quick NAT search algorithm is designed to use the Hash search instead of the sequential search to reduce latency when looking up the NAT rule table. Secondly, to leverage the power of multi-core central processing unit (CPU) and multi-queue network interface card, Quick NAT enables multiple CPU cores to process in parallel. The localized connection tracking table and the compare-and-swap based lock-free NAT Hash tables are designed to eliminate the lock overhead. Thirdly, Quick NAT uses the polling and zero-copy delivery to reduce the cost of interrupt and packet copies. The experiment results show that Quick NAT obtains high scalability and line-rate throughput on the commodity server.
Keywords:network address translation  performance optimization  Hash search  localized data structure  
本文献已被 万方数据 等数据库收录!
点击此处可从《北京邮电大学学报》浏览原始摘要信息
点击此处可从《北京邮电大学学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号