《计算机应用研究》|Application Research of Computers

软件定义网络中利用IMKVS结合NFV的分布式网络负载均衡策略

Distributed network load balancing strategy using IMKVS and NFV in software defined networks

免费全文下载 (已被下载 次)  
获取PDF全文
作者 钟百胜,姜利群
机构 1.广州工商学院 计算机科学与工程系,广州 510800;2.中国矿业大学 计算机学院,江苏 徐州 221116
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)05-047-1504-06
DOI 10.19734/j.issn.1001-3695.2018.01.0001
摘要 社交网络和其他云应用程序应该能对从数据中心发出的请求作出快速响应,实现这种请求的技术之一是内存中的键值存储(IMKVS),它是一种缓存机制,目的是为了提高整体用户体验。IMKVS系统使用一致性哈希来决定在何处存储目标。一致性哈希使用起来方法简单,但可能引起网络负载的不平衡。为了提高IMKVS的缓存性能,提出一种软件定义网络中利用IMKVS结合NFV的分布式网络负载均衡策略。该策略包含两个阶段:第一阶段设计通用的SDN负载平衡器的模块,以运行不同的负载平衡算法;第二阶段是基于IMKVS的专业化缓存,可以实现通信管理和数据复制。仿真结果表明,相比于一致性哈希,缓存服务器上的负载可改善24%,网络上的负载可改善7%,策略能够使资源利用更合理,获得更好的用户体验。
关键词 软件定义网络; 内存中的键值存储; 网络功能虚拟化; 负载均衡; 缓存; 应用层流量优化
基金项目 国家自然科学基金资助项目(61762086)
广东省青年创新人才类基金项目(2017KQNCX233)
本文URL http://www.arocmag.com/article/01-2019-05-047.html
英文标题 Distributed network load balancing strategy using IMKVS and NFV in software defined networks
作者英文名 Zhong Baisheng, Jiang Liqun
机构英文名 1.Dept. of Computer Sciences & Engineering,Guangzhou College of Technology & Business,Guangzhou 510800,China;2.College of Computer Science,China University of Mining & Technology,Xuzhou Jiangsu 221116,China
英文摘要 Social networks and other clouding applications should require fast responses from datacenter. In-memory key-value storage(IMKVS) is one of the techniques to achieve such requirement, and it is a caching mechanisms for improving overall user experience. Commonly, IMKVS systems use consistent hashing(CH) to decide where to store an object, which may cause network load imbalance due to its simplistic approach. In order to improve IMKVS performance, this paper proposed a distributed network load balancing strategy using IMKVS and NFV in software defined networks. The strategy consisted of two phases. In the first phase, it designed a generic SDN load balancer module to run different load balancing algorithms. The second phase was a specialized caching based on IMKVS to be enabled communication management and data replication. Simulation results show there is an improvement of 24% on the load of the caching servers and 7% on the load of the network compared to consistent hashing, which results in better resource usage and better user experience.
英文关键词 software defined networks(SDN); in-memory key-value storage(IMKVS); network function virtualization(NFV); load balancing; cache; application-layer traffic optimization
参考文献 查看稿件参考文献
 
收稿日期 2018/1/4
修回日期 2018/2/9
页码 1504-1509
中图分类号 TP393
文献标志码 A