Ant colony optimization load balancing routing algorithm based on SDN

Abstract

The invention request to protect an ant colony optimization load balancing routing algorithm based on a SDN. The ant colony optimization load balancing routing algorithm comprises the following steps; firstly acquiring each link state information in a SDN network to establish a load balancing model, wherein the link bandwidth capacity is limited in the load balancing module, and the limitation on the switch flow table capacity is further considered; and then putting forward an ant colony optimization algorithm solving load balancing model, wherein the proposed ant colony optimization algorithm mainly comprises selecting the next node according to a probability formula and judging whether the next node is a destination node; performing update setting on an information element update rule on the corresponding link when ending once circle, and solving the current optimal path set and outputting the same until the updated rule satisfies an iteration end condition. By use of the ant colony optimization load balancing routing algorithm disclosed by the invention, the traffic in the network can be uniformly distributed on the links, and the insufficient switch flow table space caused by excessive flow table rules is effectively avoided at the same time.

Description

technical field [0001] The invention belongs to the load balancing routing technology in the SDN network, and proposes an SDN-based ant colony optimization load balancing routing algorithm. Background technique [0002] Under the SDN architecture, the logically centralized controller can obtain the basic information of the switch and the path, calculate the appropriate data flow path according to the operator's intention and the basic information of the network, and then deliver the corresponding forwarding rules to the corresponding switch. The forwarding rules in the SDN switch are stored in a limited-sized ternary content addressable memory (TernaryContent Addressable Memory, TCAM), which supports fast parallel query of wildcard patterns, it can only save a few thousand rules, because it is very expensive and power consumption. The practical importance of limiting the size of the flow table has been recognized by the ASCII industry. Existing commercial OpenFlow switches...

AI Technical Summary

Problems solved by technology

This scheme ensures that the priority flow table in the network does not exceed its capacity by limiting the maximum number of paths of nodes, but it cannot guarantee the link load balance in the network, so it is difficult to avoid network congestion.

[0006] According to relevant research, the current load balancing scheme mainly considers the load of the link, and the phenomenon of unsuccessful data flow transmission in the network may be caused by the limited flow table of the switch.

Images