Service Abnormality Detection with Business-Aware Performance Indices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abnormality determination systems fail to accurately identify system abnormalities due to reliance on hardware-based indices, neglecting business-specific circumstances and timing, leading to unnecessary notifications or missed detections.
Innovation Solution
An abnormality determination system that acquires performance indices from a service providing system, using these to determine abnormalities based on business-specific criteria, including campaign times, and adjusts determination thresholds accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If abnormality determination is executed based solely on hardware-based indices such as response time, then the determination process is simple and fast, but the accuracy of abnormality determination deteriorates because business circumstances are not considered
Solution Approach 1:
The patent merges hardware-based indices (response time, load) with business-based indices (service utilization, campaign status) into a unified abnormality determination system. This combination allows the system to evaluate both technical performance and business context simultaneously, resolving the contradiction between simple determination and accurate determination by integrating multiple data sources into a comprehensive assessment framework.
Solution Approach 2:
The determination system is designed to handle multiple types of indices universally - both hardware metrics and business metrics can be processed through the same determination logic. This multi-functionality allows the system to adapt to different measurement types without requiring separate determination processes, maintaining simplicity while improving accuracy through diverse data consideration.
2Speed
If abnormality notification is executed immediately when load is heavy and response time is long, then rapid response is achieved, but false alarms occur when business impact is small
Solution Approach 1:
The system implements feedback by continuously monitoring both hardware indices and business indices, then using this combined information to adjust abnormality determination dynamically. The feedback loop evaluates whether observed performance degradation actually impacts business operations before triggering notifications, reducing false alarms while maintaining rapid response to genuine issues through continuous real-time assessment.
Solution Approach 2:
The abnormality determination thresholds and criteria are made dynamic rather than static. The system adapts determination standards based on current business context - for example, during campaign periods when high load is expected, the system dynamically adjusts thresholds to avoid false alarms, while maintaining strict monitoring during normal periods. This dynamic approach resolves the contradiction between rapid notification and reliable notification by making the response behavior context-dependent.
3Adaptability or versatility
If traditional hardware-based monitoring is used, then system simplicity is maintained, but the ability to detect abnormalities considering business timing and circumstances is lost
Solution Approach 1:
The monitoring system is segmented into independent modules: hardware index acquisition, business index acquisition, and determination logic. This segmentation allows business context data to be added without complicating the core hardware monitoring functionality. Each segment operates independently but contributes to the unified determination, enabling the system to gain adaptability to business circumstances while maintaining the simplicity and reliability of original hardware monitoring through modular architecture.
Data Source
AI summary
An abnormality determination system, comprising at least one processor configured to: acquire, based on a utilization situation about a service providing system for providing a predetermined service, a performance index about performance of the predetermined service; and execute, based on the performance index, abnormality determination with regard to an occurrence of abnormality in the service providing system.