Multi-constrained QoS (Quality of Service) routing strategy designing method for software defined network

Abstract

The invention discloses a multi-constrained QoS (Quality of Service) routing strategy designing method for a software defined network. The multi-constrained QoS routing strategy designing method is implemented through three modules which comprise a traffic monitoring module, a QoS routing calculation module and a blocking traffic scheduling module, wherein the traffic monitoring module records bandwidth information, port load information and detailed information of issued traffic of a link; the QoS routing calculation module uses the bandwidth and port load information, which is acquired by the traffic monitoring module, of the link to act as a routing selecting indicator, and calculates an optimal path conforming to multiple constraint conditions by using an optimization algorithm; and a controller carries out precise control through the blocking traffic scheduling module when congestion occurs at a port. According to the invention, a path calculation function based on a multi-path ant colony algorithm in an SDN (software defined network) controller is realized; and meanwhile, by adopting the routing strategy designing method which combines routing calculation and blocking scheduling, the utilization rate of the link is effectively improved, and the network load is effectively balanced.

Description

technical field [0001] The invention belongs to the technical field of computer networks, and in particular relates to a software-defined network-oriented multi-constraint QoS routing policy design method. Background technique [0002] In the current SDN network environment, the path calculation module of the Floodlight controller uses the Dijkstra algorithm to calculate the shortest path. The road is idle and cannot be fully utilized. Therefore, it is very important to study the multipath routing strategy in the SDN environment. Multipath routing selection needs to consider factors such as bandwidth, delay, packet loss rate, delay jitter, and cost. These factors are used to measure network QoS (Quality Of Service (QoS) is the basic performance index, so the multi-path routing problem is to find the optimal solution under multiple constraints. [0003] However, since there is no unified measurement scale for multi-constraint conditions, the amount of calculation to find th...

AI Technical Summary

Problems solved by technology

[0003] However, since there is no unified measurement scale for multi-constraint conditions, the amount of calculation to find the optimal routing solution under multi-constraint conditions in the traditional network environment is very large, making target optimization very difficult. In the SDN environment, the controller can obtain global The network topology view is convenient for centralized control of the entire network. Before path calculation, links that do not meet QoS constraints are excluded from the path selection object. The calculation cost of making full use of network status information to find the optimal path will be greatly reduced. However, The Dijkstra algorithm used in existing SDN controllers is only suitable for small-scale networks or laboratory networks with small traffic. The requirements for delay and link utilization are not high, and generally pass within the physical upper limit of bandwidth. The requirements of normal communication can be met by increasing the link bandwidth, but for networks with large traffic, high delay requirements, and no redundant links, the Dijkstra algorithm cannot meet the requirements. When large-scale data appears in the network When the data transmission rate reaches the maximum bandwidth of the path, the data transmission delay will increase rapidly, and even the data packet will be lost.

Images