Compression Search Control for Lower-Latency Token Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current lossless data compression algorithms, such as LZ77, face inefficiencies in compression and decompression processes due to the need to search for repeated patterns across the entire input stream, leading to increased latency and resource utilization in data centers, cloud, and networking environments.
Innovation Solution
The proposed solution involves a compression engine that selectively halts search units from searching for character string matches once a token has been identified, allowing for power conservation and resource allocation to other parts of the input stream, using heuristics like search string match length, distance offset, and literal length to determine when to skip searching, and employing a Deallocation and Delayed Match Mode to optimize the selection of tokens and literals for inclusion in the encoded stream.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If search units continuously search for character string matches across the entire input stream, then compression ratio is improved, but encoding latency and resource utilization increase
Solution Approach 1:
The patent applies partial action by selectively halting search units from searching for character string matches once a token has been identified. Instead of continuously searching across the entire input stream, the system performs searches only in necessary regions, stopping early when matches are found or when resources need reallocation. This reduces encoding latency while maintaining acceptable compression ratios by balancing search thoroughness with resource constraints.
Solution Approach 2:
The patent implements dynamic resource allocation where search units can be dynamically allocated and deallocated based on current processing needs. The system adjusts the number and activity of search units during encoding, allowing flexible adaptation to varying input characteristics and resource availability, thereby optimizing the trade-off between compression quality and encoding speed.
2Productivity
If multiple search units operate simultaneously to find all possible matches, then compression efficiency is improved, but power consumption and resource utilization increase
Solution Approach 1:
The patent employs periodic action through Deallocation and Delayed Match Mode, where search units are periodically allocated and deallocated based on processing stages. Instead of maintaining continuous operation of all search units, the system activates search units only when needed for specific matching tasks, then deallocates them afterward. This periodic activation pattern reduces power consumption while maintaining compression efficiency through staged processing.
3Measurement precision
If search units search the entire input stream for all possible character string matches, then token selection accuracy is improved, but resource utilization and processing overhead increase
Solution Approach 1:
The patent extracts and processes only the most relevant character string matches rather than analyzing the entire input stream. By using heuristics to identify and extract promising match candidates, the system focuses computational resources on the most likely token selections, reducing processing overhead while maintaining token selection accuracy through targeted analysis of critical regions.
Data Source
AI summary
In connection with compression of an input stream, multiple portions of the input stream are searched against previously received portions of the input stream to find any matches of character strings in the previously received portions of the input stream. In some cases, matches of longer character strings, as opposed to shorter character strings, can be selected for inclusion in an encoded stream that is to be compressed. Delayed selection can occur whereby among multiple matches, a match that is longer can be selected for inclusion in the encoded stream and non-selected a character string match is reverted to a literal. A search engine that is searching an input stream to identify a repeat pattern of characters can cease to search for characters that were included in the selected character string match.