Log Entry Compression via Image Seam Carving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In cloud environments, centralized logging systems face challenges with processing and scalability due to large volumes of logs being transmitted across multiple routers/nodes, leading to resource consumption and information loss, especially when brevity control is not well-coordinated across different hops, resulting in dropped useful log entries and duplicated logs.
Innovation Solution
The method involves converting logging entries into images using pixel values related to their content, applying image processing techniques like seam carving to reduce image size, and then converting the reduced image back to log entries, thereby reducing the number of logging entries without losing important information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If all logs are collected and transmitted to centralized space, then complete log information is available for analysis, but resource consumption increases and information loss occurs due to uncoordinated brevity control
Solution Approach 1:
The patent applies preliminary action by performing log reduction at the source before transmission. The log reduction unit converts logging entries to images and reduces image size before the logs leave the computing device, preventing unnecessary data from being transmitted across the network. This resolves the contradiction by maintaining information completeness for important logs while reducing overall resource consumption through pre-processing.
Solution Approach 2:
The patent introduces an intermediary mechanism - the image conversion and reduction process - between the original log data and the transmitted data. By converting logs to images and then reducing those images, the system creates an intermediate representation that preserves essential information patterns while significantly reducing data volume, thus balancing information completeness with resource efficiency.
2Adaptability or versatility
If log reduction is performed without predefined filter rules, then adaptability to dynamic environments improves, but risk of dropping useful log entries increases
Solution Approach 1:
The patent applies parameter changes by transforming logs from text format to image format with varying pixel densities. The image conversion process changes the parameters of log representation, allowing the system to adapt to different log types and environments without predefined filters. The reduction process further adjusts parameters by modifying image resolution while preserving essential patterns, achieving adaptability without sacrificing reliability.
Solution Approach 2:
The patent replaces the mechanical system of predefined filter rules with an image-based representation system. Instead of using text-based filtering mechanisms that require explicit rules, the system converts logs to images where important information is preserved through visual patterns. This substitution enables adaptability to dynamic environments while maintaining reliability through the visual preservation of log patterns.
3Productivity
If image size is reduced using compression algorithms, then data transmission efficiency improves, but risk of losing important log patterns increases
Solution Approach 1:
The patent applies partial action by performing selective reduction on image areas containing important log patterns. Rather than uniformly compressing the entire image, the system identifies and preserves critical regions while reducing less important areas. This approach improves transmission efficiency while maintaining log pattern integrity by applying reduction selectively rather than excessively across the entire image.
Solution Approach 2:
The patent applies local quality by ensuring that different regions of the reduced image maintain appropriate detail levels. Important log patterns retain higher quality and resolution, while less critical areas are reduced more aggressively. This local differentiation of quality allows the system to improve overall transmission efficiency while preserving the integrity of important log information where it matters most.
Data Source
AI summary
Embodiments relate to reducing logging entries based on contents are disclosed. According to the embodiments, the logging entries are converted into an image with pixels whose values are related with the log content and the image size is reduced by applying an image processing technology. The reduced image is converted back to the log entries. In this way, the number of logging entries is reduced without losing important information.