面向并发业务的卫星网络服务功能链优化算法  

Optimization Algorithm of Satellite Network Service Function Chain for Concurrent Services

作　　者：潘成胜[1,2] 梁芷铭 石怀峰[1,2,3] 孔志翔 PAN Chengsheng;LIANG Zhiming;SHI Huaifeng;KONG Zhixiang(Key Laboratory of Communication and Networks,Dalian University,Dalian,Liaoning 116622,China;College of Information Engineering,Dalian University,Dalian,Liaoning 116622,China;School of Automation,Nanjing University of Science and Technology,Nanjing 210094,China)

机构地区：[1]大连大学通信与网络重点实验室,辽宁大连116622 [2]大连大学信息工程学院,辽宁大连116622 [3]南京理工大学自动化学院,南京210094

出　　处：《计算机工程》2021年第3期196-201,共6页Computer Engineering

基　　金：国家自然科学基金(61931004);国家自然科学基金青年基金(61801073)。

摘　　要：在空间信息网络中,卫星硬件在线升级难度大且并发业务数量多,导致现有服务功能链(SFC)映射算法的服务请求接收率降低且资源开销过大。利用流量缩放因子与虚拟网络功能间的依附关系构建SFC,以最小化处理请求时延与总资源消耗为目标,提出一种空间信息网络SFC映射算法。通过调整预测函数的权值来控制搜索范围,从而获得全局最优的快速映射方案,有效解决节点随机失效问题。仿真结果表明,与OMD算法相比,该算法在较高的并发服务请求下可使处理请求时延和总资源消耗平均降低19%和6%。The difficulty of online satellite hardware upgrade and the increase of concurrent services in spatial information network reduce the service request acceptance rate of the existing Service Function Chain(SFC)mapping algorithm and increase the resource overhead.Based on the SFC constructed by using the traffic scaling factor and the dependency relationship between Virtual Network Function(VNF),this paper proposes a mapping algorithm of spatial information network SFC to minimize the processing delay and resource consumption.By adjusting the weight of the prediction function to control the search range,the global optimal fast mapping scheme is obtained,which can effectively improve the random failure problem of nodes.Simulation results show that compared with OMD algorithm,the proposed algorithm can reduce the processing delay and resource overhead by an average of 19%and 6%under the higher request of concurrent services.

关 键 词：空间信息网络 网络功能虚拟化 服务功能链构建与映射 路由算法 预测函数 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]