SDN Controller Flow Table Generation for Dynamic Path Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In traditional layer-3 forwarding networks, routing protocols often fail to dynamically adjust forwarding paths based on changing traffic flow rates, leading to potential congestion and inefficiencies, as paths are typically selected based on the shortest path or ECMP without re-convergence unless failure occurs.
Innovation Solution
The implementation of a Software-Defined Networking (SDN) system that calculates multiple forwarding paths based on traffic flow rates and allocates flow rate identities to enable dynamic path selection by SDN switching devices, allowing them to choose paths based on current bandwidth demands, even if they are not the shortest or ECMP paths.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If routing protocols select paths based on shortest path or ECMP, then forwarding simplicity is improved, but network adaptability to changing traffic flow rates deteriorates
Solution Approach 1:
The patent implements dynamic path selection by enabling SDN switching devices to adjust forwarding paths in real-time based on measured traffic flow rates. The system transitions from static routing tables to dynamic flow table entries that are continuously updated according to current network conditions, allowing the network to adapt to changing traffic patterns while maintaining forwarding efficiency
Solution Approach 2:
The SDN controller acts as an intermediary between the routing protocol and forwarding devices. It receives routing information, calculates optimal forwarding paths considering traffic flow rates, and distributes flow table entries to SDN switching devices. This intermediary layer enables complex adaptive routing logic without complicating the forwarding devices themselves
2Stability of the object's composition
If routing protocols fail to dynamically adjust forwarding paths, then forwarding stability is improved, but network efficiency deteriorates due to congestion
Solution Approach 1:
The patent implements a feedback mechanism where SDN switching devices continuously measure traffic flow rates on active forwarding paths and report this information to the SDN controller. The controller uses this feedback to dynamically recalculate optimal paths and update flow table entries, creating a closed-loop system that maintains forwarding stability while adapting to congestion conditions
Solution Approach 2:
The system performs preliminary calculations of multiple candidate forwarding paths and their associated traffic flow rates before actual packet forwarding. By pre-computing alternative paths and monitoring their characteristics, the system can quickly switch to optimal paths when congestion is detected, maintaining both stability and efficiency
3Productivity
If SDN switching devices dynamically select forwarding paths based on flow rates, then bandwidth utilization is improved, but device complexity increases
Solution Approach 1:
The SDN controller serves as an external intermediary that handles the complex path calculation and flow table management tasks. Individual SDN switching devices only need to implement simple flow table lookup and forwarding operations, offloading the computational complexity to the centralized controller while maintaining high bandwidth utilization through dynamic path selection
Solution Approach 2:
The system uses flow table entries as simplified copies or representations of complex routing decisions. Instead of implementing full routing protocol logic in each switching device, the controller pre-computes and distributes simplified flow-based forwarding rules that enable dynamic path selection with minimal device complexity
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A SDN controller calculates forwarding paths for a traffic flow based on SDN topology. The controller determines a maximum flow rate and flow rate identity of each forwarding path, generates ingress flow table entries based on packet characteristic information of the traffic flow, the flow rate identity and an output port of an ingress node on each forwarding path, and sends the ingress flow table entries and maximum flow rate and flow rate identity to the ingress node; generates intermediate flow table entry for each intermediate node based on the packet characteristic information, flow rate identity and an output port of each intermediate node, and sends each intermediate flow table entry to corresponding intermediate node; generates an egress flow table entry based on packet characteristic information and an output port of an egress node for sending the traffic flow, and sends the egress flow table entry to the egress node.