Log Clustering via Source Code Location Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing log processing systems face challenges in accurately analyzing and clustering logs from multiple sources, especially when logs are encrypted or have varying location values, leading to reduced accuracy and increased complexity in identifying conditions and root causes.
Innovation Solution
A system that clusters logs based on location values and attribute values, using neural networks for classification, and includes mechanisms to handle encrypted data without decryption, allowing for accurate identification of states and conditions without compromising security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If logs are encrypted to maintain security, then system security is improved, but log analysis accuracy deteriorates
Solution Approach 1:
The patent introduces location values as an intermediary element that bridges encrypted logs and analysis systems. These location values are extracted from source code files and embedded in logs, enabling the analysis system to cluster and analyze logs based on their source locations without needing to decrypt the actual log content. This mediator approach maintains security while enabling accurate analysis.
Solution Approach 2:
The patent extracts location information from source code files and separates it from the encrypted log data. By taking out the location values and using them as standalone clustering keys, the system can perform accurate log analysis based on source code locations without compromising the encrypted nature of the actual log messages, thus resolving the contradiction between security and analysis accuracy.
2Adaptability or versatility
If logs from multiple sources are clustered together for comprehensive analysis, then monitoring coverage is improved, but false correlations increase
Solution Approach 1:
The patent segments logs into distinct clusters based on their location values that identify specific source code files and lines. By segmenting the log analysis into location-based groups, the system can comprehensively monitor multiple software objects while maintaining clear boundaries between different source locations, thus preventing false correlations between logs from different sources.
Solution Approach 2:
The patent applies local quality by making each log cluster have specific characteristics based on its source location. Each cluster is associated with particular source code files and lines, giving local identity and context to groups of logs. This local differentiation allows comprehensive multi-source monitoring while maintaining high correlation accuracy within each location-specific cluster.
3Measurement precision
If detailed location values are included in logs to improve traceability, then identification precision is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary action by extracting and embedding location values into logs at the point of log generation, before the logs need to be analyzed. The system pre-processes source code files to identify relevant location information and attaches it to logs as they are created. This preliminary preparation simplifies subsequent analysis operations while maintaining high location identification precision.
Data Source
AI summary
A system for processing logs comprising at least one computing device adapted to: receiving a plurality of historical logs each generated by at least one client computing device when executing at least one software object generated from at least one source code file, each of the plurality of historical logs is associated with one of a plurality of sessions and comprising log data and a location value identifying a location in the at least one source code file; generating a plurality of log clusters, each log cluster comprising a plurality of logs of the plurality of historical logs, each of the plurality of logs of the respective log cluster comprising a common location value; and in at least one of a plurality of iterations: receiving from the at least one client computing device, via at least one digital communication network interface connected to the at least one computing device.