Router PDU Compression via Frame Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional compression algorithms are inefficient for data packets or frames in high data rate networks, leading to increased latency and reduced network throughput due to the need to aggregate frames before compression.
Innovation Solution
A method and system for compressing and decompressing network data at a packet or frame level, where a local router determines whether to compress a source PDU, sets a flag indicating compression, and transmits the compressed PDU via a network link, using a compression/decompression module that manages buffers and checksums to enable real-time compression without aggregating frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If conventional compression algorithms aggregate frames in a buffer before compression, then compression efficiency is improved, but latency increases and network throughput decreases
Solution Approach 1:
The patent divides the data stream into individual frame-level segments rather than aggregating large buffers. Each frame is compressed independently using LZLP algorithm, allowing real-time compression without waiting for buffer aggregation, thus reducing latency while maintaining compression efficiency at the frame level
Solution Approach 2:
The patent pre-allocates small output buffers for each frame compression operation and pre-initializes the compression state machine. This preliminary preparation eliminates buffer management overhead during actual compression, enabling real-time processing without latency penalties
2Loss of substance
If conventional compression algorithms use large input buffers for big data chunks, then compression ratio is improved, but processing speed decreases
Solution Approach 1:
The patent segments the data processing into frame-level units rather than large buffer chunks. Each frame is compressed independently with dedicated small buffers, enabling parallel processing and maintaining high network throughput while achieving effective compression ratios through the LZLP algorithm's frame-level optimization
Solution Approach 2:
The patent changes the buffer size parameter from large aggregation buffers to small frame-level buffers, and adjusts the compression algorithm parameters to operate optimally at frame level. This parameter transformation enables both high compression efficiency and high processing speed by matching buffer dimensions to network frame structures
3Loss of time
If compression is performed at frame level without aggregation, then latency is reduced and throughput is improved, but compression efficiency decreases
Solution Approach 1:
The patent uses a token-based copying mechanism where compressed frame representations are stored and reused. The LZLP algorithm creates compact token sequences that can be efficiently copied and referenced, maintaining high compression ratios at frame level by eliminating redundant data patterns within each frame without requiringè·¨-frame aggregation
Solution Approach 2:
The patent optimizes compression parameters specifically for frame-level operation, adjusting dictionary sizes, token buffer dimensions, and compression thresholds to maximize efficiency at the individual frame level. These parameter changes enable the algorithm to achieve compression ratios comparable to or better than buffer aggregation methods while processing frames independently in real-time
Data Source
AI summary
Method and system for processing a source protocol data unit (PDU) is provided. The method includes receiving the source PDU at a local router; determining whether to compress the source PDU or process the source PDU without compression; compressing the source PDU and setting a flag indicating that the source PDU is compressed; and transmitting the compressed source PDU via a network link. The system includes a host system that generates a source PDU; a local router that receives the source PDU and determines whether to compress the source PDU or process the source PDU without compression; compresses the source PDU and sets a flag indicating that the source PDU is compressed; and transmits the compressed source PDU via a network link to a remote router.