Multi-path load balancing method based on price mechanism for industrial real-time network

Abstract

The invention discloses a multi-path load balancing method based on the price mechanism for an industrial real-time network. The method specifically comprises the following steps: an RYU control sendsa data packet to sense an overall network topology; real-time bandwidth and delay monitoring is carried out through the OpenFlow protocol; k shortest paths from feasible paths are calculated throughthe KSP algorithm and used as candidate path; link evaluation is carried out on the k candidate paths, the candidate path with the optimum current link status serves as a forwarding path, and the candidate path with the second optimum current link status serves as a reserved path; the RYU controller sends a forwarding strategy for a current network flow to flow tables of corresponding switches; and the RYU controller carries out real-time link status monitoring, and when a link utilization rate is larger than a threshold, reselects the optimum path other than the current path for rerouting andsends the flow tables to the switches. The method provided by the invention has the advantages that the network resource utilization rate can be improved, and network congestion can be reduced; and an SDN controller is used for whole network monitoring on a data center, and the forwarding path is selected according to current real-time network conditions, so that dynamic multi-path load balancingcan be achieved.

Description

technical field [0001] The invention relates to an industrial network environment, in particular to a multi-path load balancing method based on a price mechanism in an industrial real-time network Background technique [0002] The Internet actually operates with the support of countless data centers, which provide the necessary intelligent processing and information storage functions for the Internet. After decades of development, the data center has become an economic infrastructure like transportation and energy. In recent years, with the rapid development of emerging technologies such as virtualization and cloud computing, the popularization of smart phones and 4G networks, and the rapid growth of demand for bandwidth and traffic in various industries, the data center market is expanding rapidly. In 2017, China The overall market size of IDC is expected to exceed 90 billion. According to the data from the China IDC circle, the scale of China's IDC market will maintain a...

AI Technical Summary

Problems solved by technology

When low-level equipment carries high-load traffic, it is very easy to cause problems such as packet loss, long delay, and high CPU, which affects network performance.

And the purchase of advanced equipment with higher configuration has greatly increased the cost

[0006] (3) After the network scale increases, the reliability and scalability of the two-layer network become new bottlenecks

However, the path selected by STP may not be the optimal path in the current network state, and there are problems such as slow convergence time, high pressure on the root bridge, and waste of redundant link resources.

[0007] (4) The functions of traditional network equipment are defined by the manufacturer, and administrators can only configure and apply them according to the existing functions

[0010] In an industrial environment, the existing hardware technology can support the link to transmit information at a certain low load rate. Therefore, through price adjustment, it is a worthwhile method to limit the utilization rate of the link with a high price.

Images