Virtual Machine Migration Across Wide Area Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for migrating virtual machines between distributed computing networks across a wide area network are inefficient due to high data transfer times and downtime, as they often require pausing applications and utilize suboptimal bandwidth, especially when transferring disk and memory states.
Innovation Solution
The implementation of a distributive computing network manager that establishes a data link across a wide area network, allowing for asynchronous and synchronous transfer of virtual machine states, using optimizations such as content-based redundancy, page deltas, and iterative memory page transfers to reduce data transfer volume and downtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If virtual machines are migrated across a wide area network using traditional methods, then the migration can be completed, but the data transfer time is excessively long and the virtual machines experience extended downtime
Solution Approach 1:
The system performs preliminary actions by establishing the data link and preparing the destination environment before actually migrating the virtual machine. The source virtual machine is paused only after the destination is ready, minimizing downtime. This includes pre-configuring the destination host, establishing network connectivity, and preparing storage resources before the migration process begins.
Solution Approach 2:
The migration process is segmented into distinct phases: preparation phase (establishing data link, configuring destination), migration phase (transferring virtual machine components), and completion phase (switching traffic, cleanup). This segmentation allows for optimized handling of each phase, reducing overall migration time and improving productivity.
2Loss of energy
If all virtual machine data is transferred during migration, then the migration is complete, but the bandwidth consumption is excessive and transfer time is prolonged
Solution Approach 1:
The system extracts only the essential and changed components of the virtual machine for transfer, rather than moving the entire virtual machine image. This includes extracting only modified disk blocks, active memory pages, and necessary configuration data. By taking out only what is needed, bandwidth consumption is reduced and transfer time is shortened.
Solution Approach 2:
The system performs partial action by transferring only the necessary portions of virtual machine data rather than complete copies. It uses incremental updates and delta transfers to move only changed data blocks, avoiding excessive data transmission. This partial action approach significantly reduces bandwidth consumption while maintaining migration completeness.
3Reliability
If the virtual machine is paused during migration to ensure data consistency, then data integrity is maintained, but service availability is reduced
Solution Approach 1:
The system performs preliminary actions to establish the destination environment and data link before pausing the source virtual machine. This ensures that when the pause does occur, the destination is ready to immediately receive and activate the virtual machine, minimizing service disruption. The pause is brief because all preparatory work has been completed in advance.
Solution Approach 2:
The system uses feedback mechanisms to monitor the migration progress and data consistency. It continuously checks the state of the virtual machine and adjusts the migration process accordingly, allowing for minimal pausing while maintaining data integrity through coordinated synchronization between source and destination systems.
Data Source
AI summary
Methods and apparatus to migrate virtual machines between distributive computing networks across a network are disclosed. A disclosed example method includes establishing a data link across a network between a first distributive computing network and a second distributive computing network, the first distributive computing network including a virtual machine operated by a first host communicatively coupled to a virtual private network via a first virtual local area network, communicatively coupling a second host included within the second distributive computing network to the virtual private network via a second virtual local area network, and migrating the virtual machine via the data link by transmitting a memory state of at least one application on the first host to the second host while the at least one application is operating.