Network Node Unit Hash Partitioning for Routing Lookup Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communications networks face challenges in efficiently handling packets due to limitations in packet processing capabilities and the complexity of routing systems, especially with the proliferation of IPv6 and IoT devices.
Innovation Solution
A method and system that utilize a network node unit to read routing entries, mask network addresses, and perform a hash process to associate them with forwarding units, thereby partitioning routing entries and reducing the lookup volume in forwarding information bases.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If routing information is stored in a large FIB table with varying prefix lengths, then routing completeness is improved, but lookup speed deteriorates
Solution Approach 1:
The patent divides the large FIB table into multiple smaller FIB tables, each managed by a different forwarding element. This segmentation reduces the lookup volume in each individual table, improving lookup speed while maintaining complete routing coverage across the distributed system.
Solution Approach 2:
The patent introduces a hash-based distribution mechanism that maps routing entries to forwarding elements using a hash function. This adds a dimensional transformation from direct table lookup to hash-based indirect access, enabling faster lookup performance while preserving routing completeness.
2Productivity
If multiple forwarding elements with own hardware resources are used, then forwarding capacity is improved, but system complexity increases
Solution Approach 1:
The patent implements a universal hash function that can be applied across all forwarding elements regardless of their specific hardware configurations. This universal approach simplifies the system by providing a consistent distribution mechanism while still enabling each forwarding element to utilize its own hardware resources for parallel processing.
Solution Approach 2:
The patent introduces a control element as an intermediary that manages the distribution of routing information to forwarding elements. This intermediary coordinates the complex interactions between multiple forwarding elements, simplifying the overall system architecture while maintaining high forwarding capacity.
3Speed
If per-packet fast lookup is implemented in large FIB tables, then forwarding speed is improved, but processing complexity increases
Solution Approach 1:
The patent segments the large FIB table into smaller distributed tables, reducing the processing complexity required for each individual lookup operation. This segmentation enables fast per-packet lookup by limiting the search space in each forwarding element's local table.
Solution Approach 2:
The patent replaces the mechanical sequential search through a large FIB table with a hash-based direct access mechanism. This substitution transforms the lookup process from a complex sequential comparison operation to a simpler hash computation and direct table access operation.
Data Source
AI summary
According to embodiments herein e.g. a method performed by a network node unit for managing packet handling in a communications network is provided. The network node unit reads through routing entries in a routing information base. The network node unit performs for one or more routing entries, a hash process on a network address of the routing entry into a hashed network address associated with an identity of a forwarding unit. The network node unit sends information relating to the routing entry to the forwarding unit with the identity thereby partitioning the routing entries to different forwarding units e.g. targeted by a correlated traffic distribution mechanism.