SDN Controller Adaptive Path Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software-defined networking (SDN) controllers assume infinite flow entries, leading to switch overloading and potential misidentification of the fastest data path, as the shortest path may not always be the fastest due to excessive flow entries.
Innovation Solution
A method to calculate the shortest data path by receiving information on the maximum capacity of flow entries from each switch, specifying a threshold, identifying switches at or near capacity, and excluding them from the path calculation to optimize data packet forwarding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the SDN controller assumes infinite flow entries capacity for all switches, then the path calculation is simplified and faster, but switches become overloaded and the identified shortest path may not be the fastest actual data path
Solution Approach 1:
The SDN controller performs preliminary actions by obtaining flow entry capacity information from switches in advance and calculating threshold values before performing path calculations. This allows the controller to proactively identify and exclude switches that would become overloaded, ensuring that subsequent path calculations automatically avoid problematic switches without adding complexity to the calculation process itself.
Solution Approach 2:
The patent applies local quality by treating switches with different flow entry capacities differently. Instead of using a uniform assumption for all switches, the controller obtains specific capacity information for each switch and calculates individualized threshold values. This allows the path calculation to adapt to local conditions at each switch, excluding only those switches that would become overloaded while maintaining others in the calculation.
2Reliability
If the SDN controller monitors and excludes switches near flow entry capacity, then data path performance is improved by avoiding overloaded switches, but the controller complexity and computational overhead increase
Solution Approach 1:
The patent changes the parameter used in path calculation from a static assumption of infinite capacity to a dynamic threshold value based on actual switch capacity. The controller obtains flow entry capacity information from switches and transforms this into threshold values that are then used to filter switches during path calculation. This parameter change enables adaptive path selection without requiring complex real-time monitoring of actual flow entry usage.
Solution Approach 2:
The patent introduces an intermediary mechanism in the form of threshold values that mediate between the SDN controller and the actual switch capacity. Instead of directly monitoring and managing flow entries on switches, the controller uses these calculated thresholds as an intermediary layer to indirectly control and optimize path selection. This intermediary approach simplifies the controller's role while still achieving performance optimization.
3Productivity
If the SDN controller uses the shortest path algorithm without considering switch capacity, then routing decisions are made quickly and simply, but flow entries may be installed on switches that cannot accommodate them, causing overloading
Solution Approach 1:
The SDN controller performs preliminary action by obtaining flow entry capacity information from switches before making routing decisions. This advance information gathering allows the controller to pre-calculate threshold values that reflect actual switch capacities. When performing path calculations, the controller simply excludes switches below their thresholds, maintaining fast routing decisions while preventing capacity violations from the outset.
Data Source
AI summary
Embodiments of the present invention provide methods, computer program products, and systems for generating a shortest data path for data packets. Embodiments of the present invention can be used to exclude switches at or near their maximum capacity of flow entries from the shortest data path calculation. Embodiments of the present invention can be used to reduce the “lag time” users can experience while waiting for data packets to be forwarded through different switches.


