计算机科学与探索 ›› 2018, Vol. 12 ›› Issue (7): 1109-1116.DOI: 10.3778/j.issn.1673-9418.1706022

• 网络与信息安全 • 上一篇    下一篇

蚁群分工启发的ICN负载均衡机制

任珂欣,王兴伟,马连博,黄敏   

  1. 1. 东北大学 计算机科学与工程学院,沈阳 110169
    2. 东北大学 软件学院,沈阳 110169
    3. 东北大学 信息科学与工程学院,沈阳 110819
  • 出版日期:2018-07-01 发布日期:2018-07-06

Ant Swarm Cooperation Inspired ICN Load Balance Scheme

REN Kexin, WANG Xingwei, MA Lianbo, HUANG Min   

  1. 1. School of Computer Science and Engineering, Northeastern University, Shenyang 110169, China
    2. Software College, Northeastern University, Shenyang 110169, China
    3. School of Information Science and Engineering, Northeastern University, Shenyang 110819, China
  • Online:2018-07-01 Published:2018-07-06

摘要:

信息中心网络(information centric networking,ICN)支持网内缓存和多路径路由,但由于路由器缓存能力有限,网络可能出现严重拥塞。针对这种情况,设计了蚁群分工协作启发的ICN负载均衡机制(ant swarm cooperation inspired ICN load balance scheme,ASCLB)。首先定期预测路由器和链路负载,再判断是否执行负载均衡机制;其次设计了蚁后表、雄蚁包和工蚁包;最后通过蚁群分工协作,寻找一条到内容服务器的新路径,将部分待处理包沿该路径迁移至轻载节点或链路。实验表明,与不考虑负载均衡的ICN路由机制相比,该机制可以有效均衡路由器、链路和内容服务器负载。

关键词: 信息中心网络(ICN), 蚂蚁分工, 蚁群算法, 负载均衡, 局部回归算法

Abstract:

Information centric networking (ICN) supports in-network caching and multi-path routing. However, due to limited cache capacity of ICN routers, network may suffer from serious congestion problems. Therefore, it is necessary to address these problems. For this purpose, this paper proposes ant swarm cooperation inspired ICN load balance scheme (ASCLB). Firstly, ASCLB predicts the load of routers and links periodically, and uses the predicted load to judge whether ASCLB needs to be executed. Secondly, queen table, drone packet and worker packet are     designed. Finally, by using ant swarm cooperation, an optimal path towards repository is returned, and pending packets are transported to underloaded router or link along migration path. The experimental results show that compared with ICN routing mechanism without load balance scheme, ASCLB can balance the loads of routers, links and repositories effectively.

Key words: information centric networking (ICN), ant swarm cooperation, ant colony optimization, load balance, local regression algorithm