Link Aggregation Hash Code Uniformity via Array Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing link aggregation methods fail to achieve even traffic distribution among links, leading to load imbalances due to limitations in hash-code functions that do not provide sufficient bits to select among all available ports, resulting in uneven traffic loads and capacity disparities.
Innovation Solution
A hash code with a substantially greater number of bits is used to select among links, employing an array populated with 3-bit port identifications to ensure random and equal distribution, even when not all links are utilized, and adjusting for link capacities to balance traffic loads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a hash-code function with limited bits (e.g., 3-bit) is used to select among links, then the device complexity is reduced, but the traffic distribution becomes uneven among links
Solution Approach 1:
The patent transitions from a direct 3-bit hash code selecting among 8 links to an 8-bit hash code selecting among 256 array entries, where each entry contains a 3-bit port identification. This dimensional expansion allows the system to achieve uniform traffic distribution across fewer links while maintaining hash code efficiency.
Solution Approach 2:
The patent introduces an intermediary array structure that maps 8-bit hash code outputs to 3-bit port identifications. This intermediary layer decouples the hash code generation from direct link selection, enabling uniform distribution while preserving the efficiency of the original hash function.
2Manufacturing precision
If a hash code with greater number of bits (e.g., 8-bit) is used to select among links, then the traffic distribution uniformity is improved, but the device complexity increases
Solution Approach 1:
The patent segments the 8-bit hash code output into multiple components: an array index (selecting from 256 entries) and a port identification (3-bit code). This segmentation allows the system to achieve uniform distribution across links while reusing the same hash code structure for different numbers of links.
Solution Approach 2:
The patent creates a universal link selection mechanism that can handle any number of links (2, 3, 4, 6, 8, or more) using the same 8-bit hash code and array structure. The system adapts to different configurations by populating the array with appropriate port identifications, eliminating the need for multiple specialized hash functions.
3Productivity
If the array is populated with port identifications in proportion to link capacities, then the bandwidth utilization is optimized, but the device complexity increases
Solution Approach 1:
The patent applies local quality by populating the array with port identifications in proportions that reflect individual link capacities. Links with higher capacity receive more array entries, directing more traffic to them, while links with lower capacity receive fewer entries. This localized adjustment optimizes bandwidth utilization without changing the overall hash code structure.
Data Source
AI summary
A system for providing a substantially balanced distribution of traffic over an aggregation of output lines carrying digital information makes use of m random or pseudo-random bits substantially greater in number than the number of bits (n) used for selection of individual lines. The m bits address a table populated with n-bit entries whose bit combinations correspond with the respective output lines, with the relative numbers of the bit combinations being such as to provide substantially equal loads on the individual lines.

