Selective Transaction Data Collection for Diagnostic Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Transaction problem diagnostic in high-volume systems is challenging due to difficulty in tracing useful data, leading to resource consumption and poor performance, as collected data is huge and complex, requiring significant analysis workload.
Innovation Solution
A data collection mechanism is implemented by determining affected transactions with exceptions, identifying trace features, generating a data collection rule based on these features, and collecting data from subsequent transactions that comply with the rule, thereby reducing unnecessary resource usage and diagnostic duration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data collection is performed on all transactions in high-volume systems, then complete diagnostic information is obtained, but resource consumption increases and system performance deteriorates
Solution Approach 1:
The patent extracts only the necessary diagnostic data from transactions by identifying affected transactions related to exceptions and collecting data only from those specific transactions rather than all transactions. This selective extraction approach maintains diagnostic completeness while reducing the volume of collected data, thereby lowering resource consumption and preserving system performance.
2Reliability
If data collection is performed on all transactions, then comprehensive diagnostic coverage is achieved, but the volume of collected data becomes huge and complex requiring significant analysis workload
Solution Approach 1:
The patent extracts only the relevant data from affected transactions by using exception information to identify which transactions are truly diagnostic-worthy. This selective extraction reduces the quantity of collected data from potentially all transactions to only those transactions that actually contain diagnostic value, while maintaining comprehensive coverage of all exception-related transactions.
Solution Approach 2:
The patent applies local quality by treating different transactions differently based on their relationship to exceptions. Instead of uniformly collecting data from all transactions, the system identifies and collects data only from transactions that are locally relevant to specific exceptions, giving each transaction a different data collection treatment based on its diagnostic importance.
3Reliability
If trace data is collected from all transactions, then complete problem diagnostic is enabled, but resource consumption increases and diagnostic duration extends
Solution Approach 1:
The patent extracts only the essential trace data from transactions that are actually affected by exceptions. By using exception information to guide data collection, the system avoids collecting trace data from unrelated transactions, thereby reducing the total amount of data that needs to be processed and analyzed, which directly reduces diagnostic time while maintaining complete problem diagnostic capability.
Data Source
AI summary
Data collection is provided, in which one or more affected transactions related to one or more transaction exceptions are determined. Based on one or more features of the one or more affected transactions, one or more trace features are determined. Based on the one or more trace features, a data collection rule is generated. Data of a subsequent transaction complying with the data collection rule is collected.