| SDN-ScaSVNE:可伸缩的SDN生存性虚拟网络映射算法 |
| |
| 引用本文： | 卢美莲,顾云,刘通. SDN-ScaSVNE:可伸缩的SDN生存性虚拟网络映射算法[J]. 北京邮电大学学报, 2018, 41(3): 75-80. DOI: 10.13190/j.jbupt.2017-213 |
| |
| 作者姓名： | 卢美莲 顾云 刘通 |
| |
| 作者单位： | 北京邮电大学 网络与交换技术国家重点实验室, 北京 100876 |
| |
| 摘 要： | 现有的生存性虚拟网络映射算法无法直接应用于软件定义网络（SDN），而且大多数算法仅通过扩展虚拟网络提供主动保护策略，性能较差.对此，提出了一种基于剩余网络资源可伸缩备份虚拟资源的SDN生存性虚拟网络映射算法，仅为扩展虚拟网络中满足备份约束的节点和链路提供备份资源，并利用剩余网络资源和备份资源共同完成故障恢复.仿真结果表明，算法在拥有较高收益/成本比值的前提下，可有效提高请求接受率和故障恢复成功率.
|
| 关 键 词： | 生存性虚拟网络映射 软件定义网络 可伸缩备份资源 |
| 收稿时间： | 2017-10-22 |
SDN-ScaSVNE: Scalable Survivable Virtual Network Embedding Algorithm in SDN |
| |
| LU Mei-lian,GU Yun,LIU Tong. SDN-ScaSVNE: Scalable Survivable Virtual Network Embedding Algorithm in SDN[J]. Journal of Beijing University of Posts and Telecommunications, 2018, 41(3): 75-80. DOI: 10.13190/j.jbupt.2017-213 |
| |
| Authors: | LU Mei-lian GU Yun LIU Tong |
| |
| Affiliation: | State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing 100876, China |
| |
| Abstract: | Existing survivable virtual network embedding algorithms cannot apply to software defined network (SDN) directly, and most of them only use active protection strategy by augmenting virtual networks, which leads to poor performance. For the problems above, a survivable virtual network embedding algorithm based on residual network resources is proposed for scalable backup virtual resources in SDN. It provides backup resources only for nodes and links that satisfy the backup constraints in the augmented virtual networks, and combines the remaining network resources with backup resources to complete failure recovery. Simulation results show that the algorithm can effectively improve the acceptance rate of virtual network requests and the success rate of failure recovery under the premise of high revenue/cost ratio. |
| |
| Keywords: | survivable virtual network embedding software-defined network scalable backup resource |
| 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 |
