Parent-Child VM Control Using Shared-Core vCPU Groups

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In nested virtualization systems, parent virtual machines lack direct control over child virtual machines, leading to management challenges and inefficiencies, particularly in distributed computing environments, as they cannot stop child VMs without modifying guest operating systems or using paravirtual interfaces.

Innovation Solution

Implement a hypervisor that supports a uniprocessor control group, where all threads within the group run on the same host processing unit, allowing parent VMs to assign priority levels and send signals to control child VMs directly, enabling efficient management without requiring modifications to the guest operating system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If parent VMs create child VMs using nested virtualization, then resource utilization improves, but control over child VMs is lost

Engineering Contradiction:
Improveresource utilizationVSAvoidcontrol over child VMs
Core Design Contradiction:
ProductivityVSEase of operation

Solution Approach 1:

The patent implements nested virtualization where a parent VM can create and manage child VMs. The parent VM acts as a hypervisor for its children, enabling hierarchical resource allocation while maintaining control through signal-based communication mechanisms between parent and child VMs.

Inventive Principle:
Principle #7Nested doll (Nesting)

Solution Approach 2:

The patent introduces control groups as an intermediary mechanism that enables the parent VM to send signals to child VMs. These control groups act as mediators that facilitate communication and control between the parent VM and child VMs without requiring direct hypervisor intervention, thus maintaining ease of operation while enabling nested virtualization.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If child VMs are allowed to run independently, then system autonomy improves, but context switching overhead increases

Engineering Contradiction:
Improvesystem autonomyVSAvoidcontext switching overhead
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent merges the parent VM and child VM into the same control group, allowing them to share the same physical processing unit. This combining reduces context switching overhead by keeping related VMs on the same core, while still maintaining system autonomy through signal-based control mechanisms that allow independent execution when needed.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent implements dynamic control where the parent VM can send signals to child VMs to control their execution. This dynamic mechanism allows the system to adapt between independent operation and coordinated control based on runtime conditions, balancing autonomy with overhead reduction.

Inventive Principle:
Principle #15Dynamics

3Measurement precision

If hypervisor manages all VMs directly, then control precision improves, but system complexity increases

Engineering Contradiction:
Improvecontrol precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the virtualization management into hierarchical levels: the hypervisor manages parent VMs with high control precision, while parent VMs in turn manage their child VMs. This segmentation reduces system complexity by distributing control responsibilities, while control precision is maintained through signal-based mechanisms that enable fine-grained control at each level.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4682723A1Techniques for managing virtual machines
Publication Date: 2026.01.21 RED HAT LLC
  • EP4682723A1 patent drawingFigure 1
  • EP4682723A1 patent drawingFigure 2
  • EP4682723A1 patent drawingFigure 3

AI summary

Systems and methods are disclosed for managing child virtual machines. An example method includes receiving, at a hypervisor running on a host computer system, a request from a parent virtual machine (VM) to create a child VM, wherein the parent VM includes a first virtual Central Processing Unit (vCPU). The method also includes creating, by the hypervisor responsive to the request, the child VM using a portion of computing resources allocated to the parent VM. The method also includes assigning a second vCPU to the child VM. The method also includes assigning the first vCPU and the second vCPU to a control group that forces the first vCPU and the second vCPU to be assigned to a same physical processing unit and gives the first vCPU of the parent VM a higher priority level.