Data-Driven Virtual Machine Recovery Path Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current VM recovery methods are inefficient in determining the fastest recovery path, leading to prolonged Recovery Time Objective (RTO) and potential business continuity disruptions due to factors like snapshot location, size, and target location variability.
Innovation Solution
A data-driven approach utilizing supervised learning to predict VM recovery times based on historical and real-time data, allowing for the selection of the fastest recovery path by training regression models with labeled data features such as snapshot size, delta objects, and network ping times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional VM recovery methods are used, then recovery can be performed from snapshot, but recovery time is prolonged due to inability to determine fastest recovery path
Solution Approach 1:
The system performs preliminary actions by collecting historical recovery data, training prediction models in advance, and pre-calculating recovery time estimates for different snapshot locations and target locations. This allows the system to quickly determine the fastest recovery path without prolonged analysis during actual recovery events.
Solution Approach 2:
The system implements feedback mechanisms by continuously collecting actual recovery time data from completed recovery operations and using this feedback to retrain and improve prediction models. This creates a closed-loop system where recovery performance data continuously enhances future recovery time predictions.
2Adaptability or versatility
If multiple snapshot locations and target locations are considered, then more recovery options are available, but determining the fastest path becomes more complex
Solution Approach 1:
The system introduces an intermediary prediction model that acts as a mediator between multiple snapshot locations and target locations. This model estimates recovery times for different path combinations, allowing the system to evaluate multiple recovery options without directly computing complex relationships between all possible paths.
Solution Approach 2:
The system changes parameters by using prediction models to estimate recovery times based on key factors such as snapshot location, target location, snapshot size, and network conditions. This transforms the complex problem of evaluating all possible recovery paths into a parameter-based estimation approach.
3Loss of time
If recovery decisions are made without predictive data, then simple decision process is maintained, but recovery time cannot be optimized
Solution Approach 1:
The system implements self-service by automatically collecting recovery data, training prediction models, and making recovery path decisions without requiring manual intervention. The system serves itself by using its own historical performance data to improve future recovery decisions autonomously.
Solution Approach 2:
The system replaces manual recovery decision-making with automated machine learning models. Instead of relying on simple rule-based systems or human judgment, the patent uses trained prediction models that automatically analyze multiple factors and recommend optimal recovery paths.
Data Source
AI summary
Techniques are provided for data-driven virtual machine restoration. In an example, a VM crashes and is to be restored. There can be multiple restoration paths that can be used to recover the VM (e.g., various source locations where a recovery image is stored, various recovery images, and various target locations where a VM can be restored to). A trained prediction model can analyze these various restoration paths and predict which restoration path will have a quickest time to recovery, to minimize a time that the VM is unavailable.