Autonomic Manager for Distributed Event Pattern Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Recent computer systems, including devices like cellular phones and home appliances, face difficulties in detecting and managing symptoms due to insufficient processing capability and storage, leading to unstable communication states and reduced operability.
Innovation Solution
A system of information processing apparatuses that detect events in a predetermined occurrence pattern by storing pattern data, generating necessary event data, selecting relevant events, and transferring them to match the pattern, allowing for distributed pattern detection and efficient symptom management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a centralized server manages symptom detection for multiple devices, then comprehensive symptom detection is improved, but communication overhead and process time increase
Solution Approach 1:
The patent divides the centralized symptom detection function into distributed components. Each information processing apparatus (device) is equipped with its own autonomic manager that can independently detect symptoms locally, while the server provides supporting functions. This segmentation eliminates the need for continuous communication for basic detection, reducing communication overhead and process time while maintaining comprehensive detection capability.
2Productivity
If each device operates an autonomic manager independently, then communication overhead is reduced, but processing capability and storage requirements increase
Solution Approach 1:
The patent implements local quality by equipping each information processing apparatus with an autonomic manager that has the specific capability to detect symptoms relevant to that device. Rather than requiring full autonomic management capability on every device, each device gets the localized functionality it needs, reducing the overall processing burden while maintaining detection efficiency.
3Reliability
If the server collects status information from all devices, then comprehensive monitoring is improved, but communication stability requirements increase
Solution Approach 1:
The patent enables each information processing apparatus to perform self-diagnosis through its local autonomic manager, which detects symptoms and generates notifications independently. This self-service capability means devices do not need to continuously communicate their status to the server for monitoring to be effective. The server receives notifications only when needed, reducing communication requirements and improving stability against communication failures.
4Device complexity
If pattern data is stored centrally on the server, then storage management is simplified, but detection speed decreases
Solution Approach 1:
The patent implements preliminary action by having each information processing apparatus store pattern data locally in its own storage device before detection is needed. This allows the autonomic manager to immediately compare events against patterns without waiting to retrieve data from the server, significantly improving detection speed. The server maintains the master copy of pattern data for updates and management purposes.
Data Source
AI summary
There is provided a system having a plurality of information processing apparatuses, each of which includes a storage device where at least one piece of pattern data indicating an occurrence pattern of events to be detected in the information processing apparatus is stored, a generation section that specifies a collection of events to be detected in the occurrence pattern based on the pattern data read from the storage device and generates necessary event data indicating the specified collection of events, a selection section that selects an event included in the necessary event data from events which have occurred in the information processing apparatus and events transferred from another information processing apparatus, and a detection section that detects if the selected event matches with the occurrence pattern indicated by the pattern data, and outputs a detection result.