Router Load Sharing via Disturbance Factor Negotiation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing router devices experience uneven multi-level load sharing, leading to potential packet loss and suboptimal network performance due to identical hash algorithms across multiple levels, resulting in high coupling degrees of load sharing results.
Innovation Solution
Each router device determines its load sharing information by negotiating with downstream devices to ensure unique disturbance factors, reducing the coupling degree of load sharing results across all levels by using a negotiation method to select distinct disturbance factors and algorithms, thereby ensuring even load distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the same hash algorithm is used by all router devices in a network, then the implementation is simple and consistent, but uneven multi-level load sharing occurs causing packet loss
Solution Approach 1:
Each router device uses the same hash algorithm (maintaining consistency) but introduces locally generated disturbance factors that are specific to each device's downstream interfaces. This allows each router to have customized load sharing characteristics while using a common algorithm, resolving the contradiction between implementation simplicity and load sharing evenness.
Solution Approach 2:
The patent modifies the hash algorithm by adding disturbance factors as a parameter change. Each router device changes the disturbance factor parameter based on its local conditions (downstream interface information), which transforms the identical algorithm into effectively different algorithms at each level, eliminating the uneven load sharing problem while keeping the base algorithm simple.
2Reliability
If disturbance factors are used to interfere with hash route selection results, then the coupling degree of load sharing results is reduced, but when network size is huge, uneven load sharing still occurs
Solution Approach 1:
The patent implements preliminary action by having router devices proactively send downstream interface information to upstream devices before actual data forwarding occurs. This advance information exchange allows upstream devices to generate appropriate disturbance factors in advance, reducing the coupling degree of load sharing results before traffic actually flows, thereby preventing uneven load sharing in large networks.
Solution Approach 2:
The patent establishes a feedback mechanism where downstream router device information is fed back to upstream devices. This feedback loop allows upstream devices to adjust their disturbance factors based on actual downstream conditions, ensuring that load sharing remains even across the entire network hierarchy. The feedback mechanism coordinates disturbance factors across multiple levels to achieve global load balancing.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
In a load sharing method and a router device, each of router devices in a load sharing relationship in a network determines load sharing information of the router device by using a negotiation method, to avoid a case in which a disturbance factor in the load sharing information of the router device is the same as a disturbance factor of another router device. Specifically, each of the router devices in a load sharing relationship obtains load sharing information of all downstream router devices that are in a load sharing relationship with the router device, and when determining load sharing information of the router device, the router device determines the load sharing information of the router device according to the load sharing information of all the downstream router devices that are in a load sharing relationship with the router device, so that a disturbance factor of the router device is different from disturbance factors of all the downstream router devices that are in a load sharing relationship with the router device. Therefore, a coupling degree of load sharing information of all the router devices in a load sharing relationship in the network is reduced, and load sharing at all levels is even.