Telecom Alarm Flood Identification and Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Telecommunications systems face issues with 'floods' of alarms, which overwhelm network management systems, leading to delays, resource exhaustion, and potential system crashes, particularly due to repeated or toggling events that exceed handling capacity.
Innovation Solution
A method and system for identifying and handling alarm floods by classifying flood events, determining preventive measures such as diverting alarms from queues, and implementing recovery actions, using a computerized approach that includes receiving, storing, and processing alarms to determine flood thresholds and classify events for appropriate handling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the network management system processes all incoming alarms, then complete monitoring coverage is maintained, but system resources are exhausted and system crash occurs
Solution Approach 1:
The patent extracts and separates flood alarms from the normal alarm processing stream by detecting alarm flood conditions and isolating these alarms into a separate handling mechanism. This allows the system to remove the harmful element (flood alarms) from the main processing queue, preventing resource exhaustion while maintaining normal monitoring operations.
Solution Approach 2:
The patent converts the harmful effect of alarm floods into a beneficial detection opportunity by monitoring alarm rates and patterns to identify flood conditions. The system uses the flood alarms themselves as triggers to activate protective measures, transforming the harmful phenomenon into a signal for system protection.
2Productivity
If the network management system increases processing speed to handle more alarms, then alarm handling capacity increases, but repeated and toggling events are not filtered, leading to resource exhaustion
Solution Approach 1:
The patent applies preliminary action by pre-configuring filters and thresholds for detecting alarm flood conditions, repeated events, and toggling events. The system prepares detection mechanisms in advance that automatically activate when flood conditions are detected, enabling rapid response without requiring high processing speed during normal operations.
Solution Approach 2:
The patent segments the alarm processing function into distinct components: normal alarm processing, flood detection, repeated event filtering, and toggling event handling. This segmentation allows each component to operate independently with optimized parameters, preventing resource exhaustion while maintaining processing capability.
3Reliability
If the system implements comprehensive alarm filtering, then resource exhaustion is prevented, but legitimate alarms may be missed, reducing monitoring accuracy
Solution Approach 1:
The patent implements dynamic filtering that adapts to different alarm conditions. The system adjusts filter sensitivity and thresholds based on real-time alarm patterns, ensuring that filtering actions are taken only when flood conditions are detected while maintaining full processing capability for legitimate alarms during normal operations.
Solution Approach 2:
The patent employs feedback mechanisms that continuously monitor alarm processing effectiveness and system resource consumption. The system uses this feedback to dynamically adjust filtering parameters, ensuring optimal balance between preventing resource exhaustion and maintaining alarm information completeness.
Data Source
AI summary
A system and method for identification and handling of a flood of alarms in a telecommunications system. A flood event occurs when a plurality of alarms in a queue of an alarm management system exceeds a threshold. The threshold may be a static, dynamic or adaptive threshold. A classification is determined for each flood event and preventive measures are determined. A plurality of floods having identical or different classifications may be identified respective of the plurality of alarms. A classification may be related, for example, to a network element (NE), element manager (EM), sync event, user actions, external application or rules. Preventive measures may be, for example, resetting a NE or EM or diverting alarms from the queue into another memory portion. The threshold may be determined respective of the total number of alarms in queue, the rate of alarms received and the rate of processing the alarms.