Hypervisor Delayed Hardware Upgrade Mode for Virtual Machines

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current virtualization systems require VM downtime for hardware changes, which can lead to security issues and complications, especially during live migration or when hypervisors do not support existing VM hardware features, and existing solutions like redundancy or hot plug techniques are either impractical or time-consuming.

Innovation Solution

Implementing a 'delayed hardware upgrade mode' in the hypervisor that queues commands for VM hardware changes to be executed during the next scheduled reboot, reducing downtime and allowing for seamless migration or load balancing by delaying incompatible hardware changes until VM reboot.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If VM hardware changes are implemented immediately, then hardware compatibility is improved, but VM downtime increases and security risks arise

Engineering Contradiction:
Improvehardware compatibilityVSAvoidVM downtime
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The hypervisor queues hardware change commands and executes them during the next scheduled VM reboot, which is a preliminary planned event. This allows hardware compatibility to be improved without causing unexpected downtime, as the reboot was already scheduled to occur soon anyway.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

A command queue acts as an intermediary between hardware change requests and their execution. The queue buffers immediate hardware change commands, allowing them to be staged and executed at an appropriate time during a scheduled reboot, thus avoiding both immediate downtime and security risks associated with ad-hoc reboots.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If redundant VM hardware is implemented, then hardware upgrade continuity is improved, but system complexity and administrative burden increase

Engineering Contradiction:
Improvehardware upgrade continuityVSAvoidsystem complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Instead of maintaining actual redundant hardware copies, the system uses virtualization to create virtual copies of hardware that can be switched instantly. The hypervisor maintains a queue of hardware configuration changes and applies them during scheduled reboots, providing upgrade continuity without physical redundancy.

Inventive Principle:
Principle #26Copying

3Adaptability or versatility

If hot plug capability is implemented, then hardware change flexibility is improved, but VM cooperation requirements and reconfiguration time increase

Engineering Contradiction:
Improvehardware change flexibilityVSAvoidVM cooperation requirements
Core Design Contradiction:
Adaptability or versatilityVSEase of operation

Solution Approach 1:

Instead of having the VM initiate or acknowledge hardware changes (hot plug approach), the hypervisor proactively queues and executes hardware changes during scheduled reboots. This inverts the control flow, eliminating the need for VM cooperation while maintaining hardware change flexibility.

Inventive Principle:
Principle #13The other way round (Inversion)

Data Source

PatentUS9965304B2Delayed hardware upgrades in virtualization systems
Publication Date: 2018.05.08 RED HAT ISRAEL
  • US9965304B2 patent drawing
  • US9965304B2 patent drawing
  • US9965304B2 patent drawing

AI summary

A system for delayed hardware upgrades in virtualization systems is disclosed. A method of the disclosure includes maintaining, by a hypervisor of a host machine, a delayed hardware upgrade mode that is applicable to virtual machines (VMs) managed by the hypervisor, receiving, by the hypervisor, a command to change VM hardware of one of the VMs managed by the hypervisor, placing, per the delayed hardware upgrade mode maintained by the hypervisor, the command in a delayed command queue associated with the VM to which the command was directed, the delayed command queue maintained by the hypervisor, and executing, by the hypervisor, the command in the delayed command queue when it is determined that the VM has been rebooted, wherein the reboot of the VM is not directly associated with the command.