Packet Identifier Generation for Network State Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for retrieving state information from packet headers in network equipment are inefficient, whether through expensive hardware implementations like content addressable memory (CAM) or slow software implementations, especially when dealing with large numbers of connections.
Innovation Solution
A scalable tester system using a field programmable gate array (FPGA) to generate and process TCP segments, with a unique packet identifier generated from a subset of packet header fields, allowing for efficient access to state memory without the need for extensive hardware resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If content addressable memory (CAM) is used for packet header lookup, then retrieval speed is improved, but cost and hardware complexity increase significantly
Solution Approach 1:
The patent extracts only the essential fields from the packet header (source IP, destination IP, source port, destination port) to form a compact lookup key. This extraction principle reduces the complexity of the lookup operation while maintaining the ability to retrieve connection state information, resolving the contradiction between fast retrieval and hardware complexity.
Solution Approach 2:
The patent changes the parameter representation by creating a condensed packet identifier from selected header fields rather than using the full header. This parameter transformation enables efficient memory addressing without requiring complex CAM hardware, achieving fast lookup with reduced hardware complexity.
2Speed
If content addressable memory (CAM) is used for packet header lookup, then retrieval speed is improved, but cost increases
Solution Approach 1:
The patent uses standard, inexpensive memory components instead of expensive CAM hardware. By implementing a software-based lookup mechanism using conventional memory, the solution achieves acceptable retrieval speed at a fraction of the cost of CAM-based solutions, directly addressing the cost issue.
3Ease of manufacture
If software sort algorithms are used for packet header lookup, then cost is reduced, but processing speed decreases
Solution Approach 1:
The patent performs preliminary actions by pre-processing packet headers to extract and combine key fields into a compact identifier before the lookup operation. This pre-processing step optimizes the subsequent memory access, improving retrieval speed while maintaining the cost-effectiveness of software implementation.
4Measurement precision
If full packet header fields are used for lookup, then connection state retrieval accuracy is improved, but lookup complexity increases
Solution Approach 1:
The patent extracts only the necessary fields (source IP, destination IP, source port, destination port) from the complete packet header to create a sufficient lookup key. This extraction maintains connection identification accuracy while dramatically reducing lookup complexity, as these four fields uniquely identify a TCP connection without requiring the entire header.
Data Source
AI summary
A method and apparatus for generating a unique packet identifier from receive packet header information accepts a series of packet words that make up a packet and selects a subset of a first packet word and a subset of a second packet word. The unique packet identifier is generated from a combination of the subsets selected from the first and second packet words for each packet received.