位置:成果数据库 > 期刊 > 期刊详情页
基于主动过载预防的移动IPv6家乡代理负载均衡
  • 期刊名称:软件学报
  • 时间:0
  • 页码:2511-2519
  • 语言:中文
  • 分类:TP393[自动化与计算机技术—计算机应用技术;自动化与计算机技术—计算机科学与技术]
  • 作者机构:[1]中国科学院计算技术研究所下一代互联网中心,北京100190, [2]中国科学院研究生院,北京100049
  • 相关基金:Supported by the National Natural Science Foundation of China under Grant No.60803139 (国家自然科学基金), the National Sciences & Technology Supporting Program of China under Grant No.2008BAH37B07 (国家科技支撑计划)
  • 相关项目:基于多家乡代理部署提升移动IPv6网络服务能力的研究
中文摘要:

在移动IPv6网络中,为确保移动节点的可寻址性,提升网络系统的服务可用性和系统性能,有必要在家乡链路部署多HA,并有效平衡负载.提出了基于主动过载预防的多HA负载均衡方法.该方法通过主动地为移动注册请求动态选择最优HA,预防HA过载,而不同于原有方案仅仅基于被动的负载迁移的思路.所提方法引入动态加权负载评估算法,为负载均衡的实施提供最佳决策依据,采用负载切片机制减小信令开销.通过理论分析证明了提出的方法比原有各方法能够更有效地减小由于HA过载造成的移动注册失败率、平均注册延时,同时引入的系统开销更小,更显著地提升MIPv6网络系统服务的可用性和系统性能.

英文摘要:

In order to enhance service availability and improve system performance, it is necessary to configure multiple HAs on the home link and efficiently balance load among these HAs. This paper proposes the Home Agent Load balancing method based on Active Overload Prevention (HALAOP). HALAOP actively selects optimal HA for each registration request to prevent HA's overload in advance, rather than just passively switching load between HAs like previous methods. HALAOP introduces a dynamic load evaluate algorithm to provide the basis for optimal load balancing decision. In addition, the load slicing mechanism is introduced to eliminate unnecessary additional overhead. The results of numeric analysis show that HALAOP can be more efficient in reducing probability of registration failure due to HA's overload and registration delay at lower additional cost than previous methods.

同期刊论文项目
同项目期刊论文