Virtualization Substrate Management for Accidental Fault Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
General virtualization systems cannot effectively utilize physical machines that experience accidental faults, as they are managed as failed machines until maintenance is completed, leading to a decrease in usable machines.
Innovation Solution
A virtualization infrastructure management device that detects accidental faults in physical machines, registers them as standby, and reallocates virtual machines from faulty machines to standby machines, allowing for the reuse of physically machines with accidental faults.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If general virtualization systems manage physical machines with accidental faults as failed machines, then system reliability is maintained, but the quantity of usable physical machines decreases
Solution Approach 1:
The patent changes the state parameter of physical machines with accidental faults from 'failed' to 'standby', allowing them to be reused after restart. This parameter change enables the system to distinguish between permanent failures and temporary faults, thereby increasing the quantity of usable machines while maintaining reliability through proper fault classification.
Solution Approach 2:
The patent introduces dynamic state management for physical machines, transitioning between operating, standby, and failed states based on fault detection. This dynamic approach allows machines to be rehabilitated and reused after accidental faults, converting a static failed state into a flexible standby state that can be activated when needed.
2Stability of the object's composition
If physical machines with accidental faults are excluded from service, then system stability is maintained, but system availability decreases
Solution Approach 1:
The patent changes the operational status parameter from 'excluded' to 'standby' for machines with accidental faults. This allows the system to maintain stability by monitoring these machines while increasing availability by making them available for service allocation, resolving the contradiction between stability and productivity.
Solution Approach 2:
The patent takes preliminary action by detecting and classifying accidental faults before they affect system availability. By proactively managing machines in a standby state after fault detection, the system prepares for potential service allocation without compromising stability, thus improving overall system availability.
3Ease of operation
If physical machines are restarted to restore from accidental faults, then machine usability is improved, but fault detection complexity increases
Solution Approach 1:
The patent enables machines to serve themselves by automatically detecting and classifying their own faults. The fault detection mechanism operates at the machine level, allowing autonomous identification of accidental faults and transition to standby state, thereby improving usability without significantly increasing detection complexity through centralized management.
Data Source
AI summary
The disclosed device makes it possible to effectively use a usable physical machine in a computer system for building a virtualization environment. A virtualization infrastructure management device 40 equipped with: a detection unit 41 for detecting the occurrence of an accidental fault which is not a physical persistent fault, in physical machine 50-i which is capable of executing virtual machine 51-i; a registration unit 42 for registering that physical machine 50-i is in standby state, in a management information storage unit 44 for storing the states of physical machines 50-1 to 50-n; and an allocation unit 43 for, when a fault occurs in physical machine 50-j, re-allocating virtual machine 51-j stopped by the fault to any of the physical machines for which the standby state is registered in the management information storage unit 44.