The virtual network embedding/mapping problem is an important issue in network virtualization in Software-Defined Networking(SDN).It is mainly concerned with mapping virtual network requests,which could be a set of SD...The virtual network embedding/mapping problem is an important issue in network virtualization in Software-Defined Networking(SDN).It is mainly concerned with mapping virtual network requests,which could be a set of SDN flows,onto a shared substrate network automatically and efficiently.Previous researches mainly focus on developing heuristic algorithms for general topology virtual network.In practice however,the virtual network is usually generated with specific topology for specific purpose.Thus,it is a challenge to optimize the heuristic algorithms with these topology information.In order to deal with this problem,we propose a topology-cognitive algorithm framework,which is composed of a guiding principle for topology algorithm developing and a compound algorithm.The compound algorithm is composed of several subalgorithms,which are optimized for specific topologies.We develop star,tree,and ring topology algorithms as examples,other subalgorithms can be easily achieved following the same framework.The simulation results show that the topology-cognitive algorithm framework is effective in developing new topology algorithms,and the developed compound algorithm greatly enhances the performance of the Revenue/Cost(R/C) ratio and the Runtime than traditional heuristic algorithms for multi-topology virtual network embedding problem.展开更多
Previous Virtual Network (VN) embedding researches mostly focus on developing heuristic algorithms to enhance the efficiency of a physical resource. However, in the equal-scale condition, where the scale of a VN is si...Previous Virtual Network (VN) embedding researches mostly focus on developing heuristic algorithms to enhance the efficiency of a physical resource. However, in the equal-scale condition, where the scale of a VN is similar to that of a substrate network, the number of successfully mapped VNs decreases sharply since bottlenecks form easily in the substrate network and disturb the embedding process. In this paper, reversed and bidirectional irrigation methods are proposed for the equal-scale and all-scale conditions. The two proposed methods can be combined with most of the existing heuristic algorithms and map a relatively large number of VNs by reducing the potential substrate bottlenecks. The simulation results show that the reversed irrigation method almost doubles the successfully mapped Revenue than the traditional one in the equal-scale condition. Meanwhile, the bidirectional irrigation method achieves the synthetically best performance in almost all scale conditions.展开更多
Service-oriented future internet architecture(SOFIA) is a clean-slate network architecture. In SOFIA, a service request is mainly processed through service resolution and network resource allocation. To realize the ...Service-oriented future internet architecture(SOFIA) is a clean-slate network architecture. In SOFIA, a service request is mainly processed through service resolution and network resource allocation. To realize the network resource allocation, we reference the idea of network virtualization and propose resource scheduling virtualization. In resource scheduling virtualization, a service request is abstracted as a virtual network(VN) and the network resources are allocated by mapping the VN onto the physical network. Resource scheduling virtualization provides centralized resource scheduling control within an autonomous system(AS) and achieves better controllability compared with the distributed schemes. Besides, resource scheduling virtualization supports multi-site selection as well. Meanwhile, we propose a collection of resource scheduling algorithms based on maximum resource tree(MRT) adapting to different scenarios. According to the simulation results, the proposed algorithms show good performance on the key metrics, such as acceptance ratio, revenue, cost and utilization. Moreover, the simulation results reveal that our algorithm is more efficient than the traditional ones.展开更多
文摘The virtual network embedding/mapping problem is an important issue in network virtualization in Software-Defined Networking(SDN).It is mainly concerned with mapping virtual network requests,which could be a set of SDN flows,onto a shared substrate network automatically and efficiently.Previous researches mainly focus on developing heuristic algorithms for general topology virtual network.In practice however,the virtual network is usually generated with specific topology for specific purpose.Thus,it is a challenge to optimize the heuristic algorithms with these topology information.In order to deal with this problem,we propose a topology-cognitive algorithm framework,which is composed of a guiding principle for topology algorithm developing and a compound algorithm.The compound algorithm is composed of several subalgorithms,which are optimized for specific topologies.We develop star,tree,and ring topology algorithms as examples,other subalgorithms can be easily achieved following the same framework.The simulation results show that the topology-cognitive algorithm framework is effective in developing new topology algorithms,and the developed compound algorithm greatly enhances the performance of the Revenue/Cost(R/C) ratio and the Runtime than traditional heuristic algorithms for multi-topology virtual network embedding problem.
基金supported by the National Basic Research Program of China under Grants No.2012CB315801,No.2011CB302901the National Science and Technology Major Projects under Grant No.2010ZX03004-002-02
文摘Previous Virtual Network (VN) embedding researches mostly focus on developing heuristic algorithms to enhance the efficiency of a physical resource. However, in the equal-scale condition, where the scale of a VN is similar to that of a substrate network, the number of successfully mapped VNs decreases sharply since bottlenecks form easily in the substrate network and disturb the embedding process. In this paper, reversed and bidirectional irrigation methods are proposed for the equal-scale and all-scale conditions. The two proposed methods can be combined with most of the existing heuristic algorithms and map a relatively large number of VNs by reducing the potential substrate bottlenecks. The simulation results show that the reversed irrigation method almost doubles the successfully mapped Revenue than the traditional one in the equal-scale condition. Meanwhile, the bidirectional irrigation method achieves the synthetically best performance in almost all scale conditions.
基金supported by the National Natural Science Foundation of China (61201153)the National Basic Research Program of China (2012CB315801)+1 种基金the Fundamental Research Funds for the Central Universities (2013RC0118)the Prospective Research Project on Future Networks in Jiangsu Future Networks Innovation Institute (BY2013095-2-16)
文摘Service-oriented future internet architecture(SOFIA) is a clean-slate network architecture. In SOFIA, a service request is mainly processed through service resolution and network resource allocation. To realize the network resource allocation, we reference the idea of network virtualization and propose resource scheduling virtualization. In resource scheduling virtualization, a service request is abstracted as a virtual network(VN) and the network resources are allocated by mapping the VN onto the physical network. Resource scheduling virtualization provides centralized resource scheduling control within an autonomous system(AS) and achieves better controllability compared with the distributed schemes. Besides, resource scheduling virtualization supports multi-site selection as well. Meanwhile, we propose a collection of resource scheduling algorithms based on maximum resource tree(MRT) adapting to different scenarios. According to the simulation results, the proposed algorithms show good performance on the key metrics, such as acceptance ratio, revenue, cost and utilization. Moreover, the simulation results reveal that our algorithm is more efficient than the traditional ones.