Cloud Service Recovery Time Prediction via User Resource Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cloud service recovery methods fail to provide personalized recovery time predictions for individual users, as they are based on overall service recovery times, which do not account for varying computing resource requirements among users, and do not address services without advance reservation mechanisms or clear prediction methods for service recovery times.
Innovation Solution
A cloud service recovery time prediction system that includes a recovery schedule storing means, resource usage profile storing means, and recovery time predicting means to identify and predict recovery times for each user's computing resources and application services based on their specific usage profiles and recovery schedules, presenting the predicted recovery time to users.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If overall service recovery time is used for all users, then system complexity is reduced, but measurement precision of recovery time for individual users deteriorates
Solution Approach 1:
The patent segments the overall service recovery time into individual computing resource recovery times. The recovery time prediction unit divides the recovery process by identifying specific computing resources (virtual machines, storage devices, networks) and their respective recovery schedules, allowing personalized recovery time prediction for each user based on their specific resource dependencies.
Solution Approach 2:
The patent applies local quality by providing different recovery time predictions tailored to each user's specific computing resource requirements. Instead of a uniform recovery time for all users, the system customizes the prediction by analyzing individual user profiles and their dependencies on specific computing resources, ensuring each user receives accurate localised recovery information.
2Measurement precision
If personalized recovery time prediction is implemented for each user, then measurement precision of recovery time improves, but device complexity increases
Solution Approach 1:
The patent implements universality through the recovery time prediction unit that serves multiple functions: identifying computing resources, retrieving recovery schedules, analyzing user profiles, and calculating personalized recovery times. This multi-functional unit handles diverse computing resources (virtual machines, storage, networks) and various user scenarios within a single integrated system, managing complexity through functional consolidation.
Solution Approach 2:
The system enables self-service by automatically predicting and notifying users of their personalized recovery times without requiring manual intervention. The notification unit autonomously sends recovery time information to users based on the prediction results, reducing operational complexity while maintaining high measurement precision.
3Loss of information
If recovery information is not provided to users, then information loss is minimized, but user anxiety increases causing system performance degradation
Solution Approach 1:
The patent implements feedback by providing users with timely and accurate recovery time predictions. The notification unit delivers personalized recovery information to users, creating a feedback loop that reduces uncertainty and anxiety. This proactive information provision prevents performance degradation by keeping users informed about service restoration timelines without causing information overload.
Data Source
AI summary
A recovery schedule storing means (81) stores a recovery schedule for a failure of a cloud service with respect to each computing resource type or each application service. A resource usage profile storing means (82) stores a resource usage profile specifying an application service or a computing resource used when each user uses the cloud service. A recovery time predicting means (83) predicts a service recovery time with respect to each user based on the resource usage profile and the recovery schedule. A recovery time presenting means (84) presents the predicted service recovery time to the user.