Multi-Access Edge Server Failure Classification for Service Reinitialization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-access edge computing environments lack effective mechanisms for managing service availability and classifying edge server failures, leading to inefficiencies in service reinitialization and impact on Quality of Service (QoS) and Quality of Experience (QoE).
Innovation Solution
Utilizing Continuous Time Markov Chains (CTMC) and Continuous Stochastic Logic (CSL) to model edge servers and service reinitialization rates, enabling context-aware classification of failures as critical or non-critical, and providing recommendations for failure criticality levels through orchestration platforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If service reinitialization is performed frequently to maintain service availability, then service availability is improved, but system performance and resource efficiency deteriorate
Solution Approach 1:
The patent changes the parameter of failure classification from binary (failure/non-failure) to multi-level (critical/non-critical) based on service availability impact. This allows differentiated reinitialization strategies where only critical failures trigger immediate reinitialization, while non-critical failures use delayed reinitialization, thus improving service availability when needed while reducing unnecessary reinitialization overhead.
Solution Approach 2:
The patent applies partial action by performing immediate reinitialization only for critical failures and delayed reinitialization for non-critical failures. This selective approach ensures that reinitialization resources are allocated only where necessary to maintain service availability, preventing excessive reinitialization that would degrade system performance.
2Measurement precision
If comprehensive failure analysis is performed to classify failures accurately, then failure classification accuracy is improved, but computational complexity and processing time increase
Solution Approach 1:
The patent segments the failure analysis process into two distinct pathways: critical failure analysis and non-critical failure analysis. Each pathway uses appropriate analysis depth and computational resources. This segmentation allows accurate classification without requiring full comprehensive analysis for all failures, thus reducing overall computational complexity while maintaining classification accuracy.
Solution Approach 2:
The patent applies different levels of analysis quality to different failure types. Critical failures receive comprehensive immediate analysis with higher computational resources, while non-critical failures use streamlined delayed analysis with reduced computational overhead. This local quality approach ensures accurate classification where needed while reducing complexity for less critical cases.
3Reliability
If immediate reinitialization is performed for all failures, then service availability is improved, but resource overhead and system stress increase
Solution Approach 1:
The patent introduces dynamic reinitialization timing based on failure criticality. Instead of static immediate reinitialization for all failures, the system dynamically adjusts reinitialization timing - immediate for critical failures and delayed for non-critical failures. This dynamic approach maintains service availability for critical services while reducing resource overhead by postponing non-critical reinitialization when resources are available.
Solution Approach 2:
The patent implements beforehand cushioning by preparing delayed reinitialization queues and resource allocation plans in advance for non-critical failures. This cushioning mechanism allows the system to handle critical failures immediately while buffering non-critical failures, reducing peak resource overhead and system stress during failure events.
Data Source
AI summary
Provided are a method, system, and computer program product in which operations are performed to model edge servers and service reinitialization rates via continuous time probabilistic models. Context-aware edge server failure classification into critical or non-critical failure is performed by analyzing a context comprising one or more services, failure rates associated with edge servers, and service reinitialization rates.