Link Aggregation Port Selection via Switch Fabric Utilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing link aggregation techniques often route network data over highly utilized paths, leading to congestion and packet loss, as they do not effectively consider the bandwidth utilization of paths through the switch fabric when selecting member ports for aggregation bundles.
Innovation Solution
The method involves evaluating the utilization of paths through the switch fabric to identify relatively low-utilization paths and applying a hashing algorithm to select member ports from a group of candidate ports, using both low and high threshold values to differentiate between lightly and moderately congested paths, thereby optimizing the use of the switch fabric and reducing congestion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If existing link aggregation techniques are used to route network data, then bandwidth utilization is achieved, but congestion and packet loss occur due to routing over highly utilized paths
Solution Approach 1:
The patent implements dynamic path selection by continuously monitoring switch fabric utilization and adapting member port selection in real-time. The system evaluates current utilization metrics and dynamically routes packets through less congested paths, transforming the static link aggregation into a dynamic system that adapts to changing network conditions, thereby preventing congestion while maintaining high bandwidth utilization
Solution Approach 2:
The patent employs feedback mechanisms by monitoring switch fabric utilization and using this information to inform subsequent routing decisions. The system collects utilization data, evaluates it against selection criteria, and adjusts member port selection based on this feedback, creating a closed-loop control system that prevents congestion and reduces packet loss while maintaining efficient bandwidth utilization
2Ease of operation
If member ports are selected without considering switch fabric utilization, then routing simplicity is maintained, but congestion occurs on highly utilized paths
Solution Approach 1:
The patent performs preliminary evaluation of switch fabric utilization before making routing decisions. By pre-assessing the utilization state of different paths and identifying less congested options in advance, the system prepares routing information that guides subsequent packet forwarding, maintaining operational simplicity while avoiding congestion through proactive path selection
Solution Approach 2:
The patent introduces an intermediary selection process that sits between the simple routing decision and the complex utilization monitoring. This intermediary layer evaluates utilization metrics and translates them into simplified member port selection criteria, maintaining ease of operation by abstracting the complexity while still achieving congestion avoidance through intelligent path selection
3Measurement precision
If utilization evaluation is performed for all paths, then optimal path selection is achieved, but computational complexity increases
Solution Approach 1:
The patent applies local quality by evaluating utilization metrics selectively for specific paths rather than uniformly across all paths. The system identifies and focuses evaluation efforts on paths that are more likely to be congested or offer better routing options, applying different levels of evaluation intensity to different parts of the routing decision space, thereby achieving accurate path selection with reduced computational overhead
Solution Approach 2:
The patent changes parameters by using threshold-based utilization metrics and categorical path classifications rather than continuous detailed analysis of all paths. By transforming continuous utilization data into discrete categories (e.g., low-utilization vs. high-utilization paths) and applying threshold comparisons, the system achieves precise path selection while significantly reducing computational complexity through parameter transformation
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This disclosure describes techniques that include selecting a member port of an aggregation bundle by evaluating utilization of paths, within a router, to member ports of an aggregation bundle. In one example, this disclosure describes a method that includes receiving network data to be output through an aggregation bundle having a plurality of member ports; identifying local member ports; identifying non-local member ports, each of the non-local member ports being reachable from the receiving line card over a path through the switch fabric to a different one of the plurality of line cards; identifying available non-local member ports by determining, for each non-local member port, whether the path through the switch fabric has low utilization; and selecting a member port by applying a hashing algorithm to a group that includes each of the identified available non-local member ports.