Nested VM Storage Snapshots Bypassing Hypervisor Exits

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current nested virtualization systems experience I/O overhead due to the need for VM exits across multiple hypervisor levels when a nested VM requests access to a virtual device, which hinders performance by requiring additional processing resources.

Innovation Solution

Assigning a snapshot of a virtual device directly to a Level 2 virtual machine, allowing it to perform storage operations without exiting to the Level 1 VM, thereby bypassing unnecessary processing steps and improving I/O efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a nested VM requests access to a virtual device in current nested virtualization systems, then the VM can access the device, but I/O overhead increases due to multiple VM exits across hypervisor levels

Engineering Contradiction:
Improvedevice access capabilityVSAvoidI/O performance
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent introduces a snapshot of the virtual device as an intermediary layer between the nested VM and the physical device. When the nested VM requests device access, it interacts with the snapshot copy rather than triggering VM exits to the Level 0 hypervisor. This intermediary snapshot handles I/O operations locally at the Level 1 hypervisor level, eliminating the performance penalty of multiple VM exits while maintaining device access functionality.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent creates a copy (snapshot) of the virtual device's storage state and assigns it directly to the nested VM. This copy serves as a local representation that the nested VM can access without requiring communication with the parent VM or the Level 0 hypervisor. The snapshot copying approach allows the nested VM to perform storage operations independently, significantly reducing I/O overhead while preserving data consistency through appropriate snapshot management.

Inventive Principle:
Principle #26Copying

2Reliability

If VM exits are required for storage operations in nested VMs, then device access is maintained through proper hypervisor control, but additional processing resources are consumed

Engineering Contradiction:
Improvehypervisor control integrityVSAvoidprocessing resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The snapshot acts as an intermediary that preserves hypervisor control integrity without requiring actual VM exits. The Level 0 hypervisor maintains control over the master copy of the virtual device, while the snapshot copy assigned to the nested VM allows local operations. This intermediary mechanism ensures that the hypervisor's control authority is preserved while eliminating the energy-consuming VM exit process for routine storage operations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent performs preliminary action by creating and assigning the snapshot of the virtual device to the nested VM before storage operations are needed. This advance preparation allows the nested VM to have immediate access to a local copy of the storage state, eliminating the need for repeated VM exits during subsequent read/write operations. The preliminary snapshot creation consumes minimal resources compared to the cumulative cost of frequent VM exits.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11983555B2Storage snapshots for nested virtual machines
Publication Date: 2024.05.14 RED HAT INC
  • US11983555B2 patent drawing
  • US11983555B2 patent drawing
  • US11983555B2 patent drawing

AI summary

Systems and methods for storage snapshots for nested virtual machines. An example method may comprise running, by a host computer system, a hypervisor managing a first virtual machine associated with a first virtual device. Responsive to creating a second virtual machine by the hypervisor, requesting, by the first virtual machine, a first snapshot of the first virtual device. The hypervisor generates the first snapshot of the first virtual device and forwards the first snapshot of the first virtual device to the second virtual machine.