Event Indication Filtering via Top-Level Class Hierarchy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In data processing systems, the excessive generation and delivery of event indications can lead to 'indication storms,' overwhelming network resources and causing unacceptably high usage levels, especially when many lifecycle or state change events occur simultaneously.
Innovation Solution
Implementing a method that detects object behavior changes in an information model and sends indications of interest only when they correspond to a specified top-level indication class in the indication class hierarchy, reducing the number of indications sent and preventing indication storms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all object behavior changes are monitored and indications are sent for every event, then complete event coverage is achieved, but network resources become overwhelmed and system performance degrades
Solution Approach 1:
The indication class hierarchy is segmented into top-level classes and subordinate classes. The patent filters indications by evaluating whether they belong to specified top-level indication classes, thereby segmenting the indication flow to prevent storms while maintaining coverage of critical events.
Solution Approach 2:
Instead of filtering out all indications except critical ones (excessive filtering), the patent applies partial filtering by sending indications only when they correspond to specified top-level classes. This partial action approach maintains sufficient event coverage while preventing network saturation.
2Loss of information
If indications are sent for every object behavior change, then all events are communicated to listeners, but the number of indications becomes excessive and causes indication storms
Solution Approach 1:
The indication class hierarchy is segmented into top-level classes and subordinate classes. The patent filters indications by evaluating whether they belong to specified top-level indication classes, thereby segmenting the indication flow to prevent storms while maintaining coverage of critical events.
Solution Approach 2:
The patent extracts only the necessary indications by checking if an indication's class corresponds to a specified top-level indication class. This extraction process removes excessive indications from the system while retaining those that are truly needed, preventing indication storms.
3Productivity
If a hierarchical filtering approach is implemented to reduce indication volume, then indication storms are prevented, but some event information may be filtered out
Solution Approach 1:
Instead of filtering out all indications except critical ones (excessive filtering), the patent applies partial filtering by sending indications only when they correspond to specified top-level classes. This partial action approach maintains sufficient event coverage while preventing network saturation.
Solution Approach 2:
The patent implements a feedback mechanism where the filter evaluates each indication against the indication class hierarchy and only blocks indications that do not correspond to specified top-level classes. This feedback loop ensures that critical events are not lost while excessive indications are filtered out.
Data Source
AI summary
A method, apparatus, and computer program product for providing event indications in an event model in a manner so that indication storms resulting from excessive indication generation and delivery is prevented. An object behavior change in an information model corresponds to an event. An indication of interest is sent to a listener process when the detected object behavior change corresponds to a specified top level indication class in an indication class hierarchy.


