Universal Data Packet Pre-Processing Apparatus for Multi-Type Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data packet pre-processing technologies are rudimentary, leaving most processing to subsequent processors and only allowing single hash functions, which limits the types of data packets that can be pre-processed, resulting in inefficient data handling and routing.
Innovation Solution
An apparatus and method for analyzing data packets by accessing and outputting information about the packet type and deriving additional information, selecting and applying specific functions based on this information, and outputting the processed data with the packet, allowing for more extensive pre-processing in hardware and improved routing of data streams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple hash functions are implemented in hardware, then more types of data packets can be pre-processed, but device complexity increases
Solution Approach 1:
The patent implements a universal pre-processing apparatus that can handle multiple data packet types (MPLS, VLAN, UDP, TCP, and other encapsulations) through a single hardware structure. This apparatus uses multiple hash functions (first hash function for MPLS/VLAN, second hash function for UDP/TCP) within one device to provide multi-functional pre-processing capabilities, resolving the contradiction between versatility and complexity by integrating diverse functions into a unified hardware platform.
2Productivity
If more processing is performed in hardware, then data handling efficiency improves, but device complexity increases
Solution Approach 1:
The patent segments the data processing workflow into distinct hardware-accelerated stages: a pre-processing apparatus that performs initial analysis and hash calculation, and subsequent processors that handle deeper packet inspection. This segmentation allows critical path operations (hash calculation for routing decisions) to be performed in hardware for high efficiency, while less time-sensitive operations remain in software, balancing productivity gains against hardware complexity.
Solution Approach 2:
The pre-processing apparatus performs preliminary actions by calculating hash values and determining routing information before packets reach the main processing queue. This preliminary hash calculation enables early routing decisions and load distribution, improving overall data handling efficiency by preparing packets in advance rather than processing everything in sequence during main processing.
3Device complexity
If a single hash function is used, then device complexity is reduced, but the types of data packets that can be pre-processed are limited
Solution Approach 1:
The patent applies local quality by selecting different hash functions based on the specific data packet type being processed. The first hash function is locally optimized for MPLS and VLAN packets, while the second hash function is locally optimized for UDP and TCP packets. This localized optimization allows each hash function to be simple and efficient for its target packet type, while the overall system maintains high versatility through conditional selection.
Solution Approach 2:
The system dynamically changes processing parameters by selecting different hash functions based on packet type identification. When an MPLS or VLAN packet is detected, the first hash function is selected; when UDP or TCP is detected, the second hash function is selected. This parameter change approach allows the system to adapt to diverse packet types without requiring a single complex hash function that handles all cases equally well.
Data Source
AI summary
An apparatus and method for analyzing a data packet, where the first and second information is derived relating to the data packet. The first information relates to a type of the data packet, a standard to which the data packet conforms and/or which data item(s) is/are present in the data packet. The first information is used for identifying a function into which the second information is input to generate third, and this third information is output together with at least part of the data packet.