Virtualization Shutdown Control via Guest OS Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In virtualization systems, shutdown processes are inefficient as they do not confirm the completion of guest OS shutdowns before proceeding, leading to potential incomplete shutdowns and wasted time, as they rely on a set time schedule rather than actual shutdown confirmation.

Innovation Solution

An information processing device with a control unit that manages the shutdown of multiple guest OSs, where each OS's shutdown status is monitored and confirmed before proceeding with the shutdown of the next OS, and the entire system is shut down only when all OSs are confirmed down, utilizing a first management unit to manage and a second unit to execute shutdowns, with an uninterruptible power supply for controlled shutdown.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a set time schedule is used for shutdown processing, then the shutdown process can proceed automatically without confirmation, but it cannot confirm whether guest OS has been shut down and may waste time or fail to properly shut down the system

Engineering Contradiction:
Improveshutdown processing speedVSAvoidshutdown completion confirmation
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The management device receives feedback information from the virtualization device indicating whether the guest OS shutdown is completed. This feedback mechanism allows the system to confirm actual shutdown status before proceeding to host OS shutdown, resolving the contradiction between automated scheduling and completion confirmation.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system performs guest OS shutdown as a preliminary action before host OS shutdown. By structuring the shutdown sequence this way and waiting for confirmation of guest OS completion, the system ensures proper shutdown order without requiring complex real-time monitoring during the actual shutdown process.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the system waits for a longer time to confirm guest OS shutdown, then shutdown completion can be confirmed more reliably, but the total shutdown time increases and system efficiency decreases

Engineering Contradiction:
Improveshutdown completion confirmationVSAvoidtotal shutdown time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Instead of waiting for a predetermined time period, the system uses feedback information from the virtualization device to determine shutdown completion. This allows the system to proceed as soon as the guest OS actually shuts down, minimizing unnecessary waiting time while ensuring reliable confirmation.

Inventive Principle:
Principle #23Feedback

3Productivity

If the system proceeds with host OS shutdown before confirming guest OS shutdown, then shutdown processing can be faster, but the guest OS may not be properly shut down and system stability is compromised

Engineering Contradiction:
Improveshutdown processing speedVSAvoidsystem shutdown stability
Core Design Contradiction:
ProductivityVSStability of the object's composition

Solution Approach 1:

The guest OS shutdown is executed as a preliminary action before host OS shutdown. The management device waits for feedback confirming guest OS shutdown completion before initiating host OS shutdown, ensuring system stability while maintaining efficient processing.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS20240427623A1Information processing device, management program, management method, and information processing system for performing shutdown of virtualization system
Publication Date: 2024.12.26 OMRON CORP
  • US20240427623A1 patent drawing
  • US20240427623A1 patent drawing
  • US20240427623A1 patent drawing

AI summary

A technique is provided for performing shutdown of a virtualization system. A control unit of a server may be configured with instructions to perform operations comprising operation as a first management unit that manages shutdown of a plurality of virtual machines that operate on a hypervisor, and a second management unit that, when all of the virtual machines are shut down, shuts down the server, and the first management unit, upon acquiring information indicating that shutdown of a virtual machine is completed, executes shutdown of another virtual machine.