Live VM Migration Using Dirty-Page Transfer for Low Downtime

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Migrating virtual machines between host machines can cause service interruptions due to the time-consuming process of copying device state information, which is undesirable for maintaining seamless user experience.

Innovation Solution

A system and method for live migration of virtual machines that involves pausing device traffic, storing dirty page information, and transferring it to a destination device within a short time frame, typically 500 milliseconds, using a control plane to configure a destination device with worker nodes and smart network interface cards to minimize downtime.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If device state information is copied from source host to destination host during virtual machine migration, then migration can be performed, but service interruption time increases

Engineering Contradiction:

Improveservice continuityVSAvoidmigration downtime

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by launching the destination device and copying non-dirty state information before pausing the source device. Worker nodes pre-configured with smart network interface cards prepare the destination environment, so that when migration occurs, only dirty pages need to be transferred, significantly reducing the actual downtime experienced by users.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The migration process is segmented into distinct phases: preliminary setup phase (launching destination, copying static state), active migration phase (pausing source, copying dirty pages), and resumption phase (transferring control). This segmentation allows parallel processing of non-critical tasks while minimizing the critical path time for state transfer.

Inventive Principle:

Principle #1Segmentation

2Loss of time

If device traffic bandwidth is limited during migration, then dirty page copying time is reduced, but overall migration speed may be affected

Engineering Contradiction:

Improvedirty page copy timeVSAvoidmigration throughput

Core Design Contradiction:

Loss of timeVSProductivity

Solution Approach 1:

The system dynamically changes bandwidth parameters during migration. By limiting device traffic bandwidth to a threshold level during the dirty page copying phase, the system ensures that migration-critical data receives sufficient network resources. This parameter adjustment prioritizes migration completion over normal traffic throughput during the transition window.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS12530219B2Time-bound live migration with minimal stop-and-copy

Publication Date: 2026.01.20 ORACLE INT CORP

AI summary

A computing system may access a source device configured to store device traffic in a source device queue. The computing system may launch, by one or more worker nodes of a control plane of the computing device, a destination device. The computing system may transmit a source device state to the destination device. The control plane worker nodes may configure the destination device from the source device state. The computing system may instruct the source device to pause storing the device traffic to the source device guest memory. The computing system may store dirty page information in the computing system memory. The computing system may transmit the dirty page information to the destination guest memory. The computing system may instruct the destination device to resume storing the device traffic to the destination guest memory.