Network Device Packet Lookup with Early Match Skipping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network devices face challenges in processing high-speed data transmission due to increased processing rates and stringent performance requirements, with existing lookup processes contributing to latency and power consumption.
Innovation Solution
Implementing a packet processor with a classification engine that performs lookups in a first database, skipping subsequent lookups in secondary databases when a match is found in the first database, thereby reducing latency and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple database lookups are performed for each packet to ensure accurate processing decisions, then processing accuracy is improved, but processing latency increases
Solution Approach 1:
The patent segments the lookup process into multiple hierarchical databases (e.g., TCAM, CAM, RAM) organized in tiers. The first database performs a quick initial lookup, and only if no match is found do subsequent databases perform additional lookups. This segmentation allows the system to achieve high processing accuracy through multiple lookups while minimizing latency by avoiding unnecessary lookups in later databases for common packet flows that match in earlier databases.
2Measurement precision
If multiple database lookups are performed for each packet to ensure accurate processing decisions, then processing accuracy is improved, but power consumption increases
Solution Approach 1:
The patent segments the lookup process into multiple hierarchical databases organized in tiers. By performing lookups in sequence through segmented databases, the system ensures processing accuracy is maintained while reducing power consumption. The segmentation allows early termination of the lookup chain for common packet flows that match in earlier databases, avoiding the energy cost of performing all possible lookups for every packet.
3Speed
If a single database lookup is performed to reduce latency, then processing speed is improved, but processing accuracy deteriorates
Solution Approach 1:
The patent segments the lookup functionality across multiple specialized databases rather than relying on a single database. Each database segment is optimized for specific types of packet classification tasks. This segmentation enables the system to achieve both high processing speed and high accuracy by distributing the lookup workload across multiple targeted databases rather than requiring one comprehensive database to handle all cases.
Solution Approach 2:
The patent adds a dimensional aspect to the lookup process by organizing databases in a hierarchical structure with multiple levels. Instead of a single-dimensional lookup approach, the system uses multi-dimensional hierarchical lookups where each level provides different types of classification information. This dimensional expansion allows the system to maintain high processing speed while achieving comprehensive processing accuracy through the layered approach.
4Adaptability or versatility
If comprehensive packet processing features are added to meet customer requirements, then functionality is improved, but device complexity increases
Solution Approach 1:
The patent segments the packet processing functionality into multiple specialized databases, each handling specific types of processing tasks (e.g., routing, quality of service, security policies). This segmentation allows the system to provide comprehensive functionality while managing device complexity by organizing features into modular, independent database segments rather than a single monolithic processing unit.
Solution Approach 2:
The patent creates a universal hierarchical lookup architecture that can handle multiple types of packet processing requirements through a single integrated system. The multi-dimensional database structure provides universal functionality for various packet classification tasks including routing, filtering, quality of service, and security policies, all through the same hierarchical lookup mechanism, thereby avoiding the need for separate dedicated systems for each function.
Data Source
AI summary
A packet processor of a network device includes a first lookup engine that generates a first lookup key for a packet, performs a first lookup in a first database using the first lookup key, and in response to finding a match of the first lookup key in the first database, determines a plurality of processing actions to be performed for the packet. Each of multiple second lookup engines selectively performs a respective second lookup in a respective second database for the packet using a respective second lookup key, and ii) selectively determines, based on the second lookup, one or more processing actions to be performed for the packet. The second lookup engines skip performing the second lookups for the packet in response to the first lookup engine finding the match of the first lookup key in the first database.