VM Migration Based on Failure Risk and Influence Range Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems fail to accurately calculate and mitigate the risk of devices being affected by failures in complex computer systems, leading to potential chain reactions and system crashes, as they lack the ability to properly determine devices at high risk before failures occur.
Innovation Solution
A system management method that calculates failure risks and influence ranges for physical and virtual devices, allowing for proactive movement of high-risk devices to low-risk physical servers to prevent failure propagation, thereby reducing the risk of system crashes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fail-over or VM movement is performed after failure occurs, then the influence of failure can be stopped from spreading, but the system cannot proactively prevent chain reaction failures before they occur
Solution Approach 1:
The patent calculates failure risks and influence ranges for all devices in advance, before failures occur. By proactively identifying devices at high risk of being affected by failures and determining optimal destination physical servers beforehand, the system enables preventive VM migration rather than reactive fail-over, thus resolving the contradiction between reliability improvement and time loss.
2Reliability
If VM is moved to another physical machine, then the analysis target device can be protected from failure influence, but the system lacks accurate information to determine the best destination
Solution Approach 1:
The patent establishes a feedback mechanism that continuously monitors operation status information of physical devices and virtual machines, updates failure risk calculations and influence range determinations based on current system state, and uses this feedback to dynamically identify high-risk devices and recommend optimal migration destinations, thereby improving both protection capability and assessment accuracy.
Solution Approach 2:
The patent replaces subjective or simplistic failure assessment methods with a systematic computational approach that uses search route information and operation status data to calculate quantitative failure risks and influence ranges, enabling precise identification of devices at high risk and data-driven destination selection.
3Reliability
If the system monitors and calculates failure risks for all devices, then proactive failure prevention is enabled, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent segments the complex system into manageable components by calculating influence ranges for individual devices based on search route information. This segmentation allows the system to analyze failure risks device-by-device rather than attempting to evaluate the entire system simultaneously, reducing computational complexity while maintaining comprehensive risk coverage.
Data Source
AI summary
A method includes: acquiring, based on status information, a failure risk of each of a plurality of devices including physical devices and virtual machines, each of the virtual machines being operated on any of the physical devices, the status information indicating the statuses of the plurality of devices; acquiring an influence range based on route information indicating a link in a range affected by a failure; acquiring a first influence risk based on a failure risk acquired for a first device, the first physical device being any of the physical devices; acquiring a second influence risk based on a failure risk of a second device, the second influence risk indicating a possibility of a target device being affected by a failure in another device; and determining the second physical devices as a destination candidate of the target device when the second influence risk is lower than the first influence risk.