Hash Table Routing for IP Flow Classification Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The latency associated with packet classification in hardware-based networking solutions using programmable network processors limits system throughput due to the resource-intensive nature of ordered lookups in ACL and PBR table lookups, particularly in scenarios where multiple packets share the same destination.
Innovation Solution
Implementing a hash table-based mechanism that utilizes historic data for predictive routing, reducing the need for repeated resource-expensive ordered lookups by caching routing information for subsequent packets with the same destination, thereby improving throughput and Quality of Service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ordered lookups in ACL and PBR tables are performed for packet classification, then routing accuracy and policy enforcement are improved, but processing latency increases and throughput decreases
Solution Approach 1:
The patent pre-calculates and stores routing information in hash tables before packets arrive. When a packet needs classification, the system first checks the hash table for pre-computed routing information based on packet flow identifiers, avoiding the need to perform resource-intensive ordered lookups in ACL and PBR tables for every packet, thus reducing latency while maintaining routing accuracy
Solution Approach 2:
The patent creates simplified copies of routing information in hash tables that contain essential routing data without the full complexity of ACL and PBR table structures. These hash table entries serve as lightweight proxies that provide fast lookup while the full ordered lookups are performed only when necessary, balancing speed and accuracy
2Measurement precision
If ordered lookups are performed for every packet to ensure accurate classification, then routing precision is improved, but system throughput decreases
Solution Approach 1:
The patent segments the packet processing workflow into two parts: a fast path using hash table lookups for packets with identical flow identifiers, and a slow path using ordered lookups in ACL and PBR tables for packets requiring detailed classification. This segmentation allows most packets to be processed quickly while maintaining precise classification when needed
Solution Approach 2:
The patent applies partial action by performing only hash table lookups for packets that can be classified using cached routing information, avoiding the excessive resource consumption of full ordered lookups. The system performs the minimal necessary action (hash lookup) for common cases while reserving full classification for exceptional cases
3Speed
If hash table caching is implemented for predictive routing, then processing speed is improved, but memory requirements increase
Solution Approach 1:
The patent implements local quality by creating hash table entries only for specific packet flow identifiers that are actively being processed, rather than maintaining comprehensive routing tables for all possible packets. The hash table stores localized routing information relevant to current traffic patterns, optimizing memory usage while maintaining fast processing for cached flows
Data Source
AI summary
Aspects of the disclosure pertain to a system and method for providing prediction based, fast routing of IP flows. A hash table-based mechanism is implemented by the system such that classification information obtained and/or utilized for a first packet of an IP flow is applied to subsequent packets of the IP flow, thereby promoting packet processing efficiency for the flow.