Virtual Machine Migration Sequencing Under Resource Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual machine migration technologies are inadequate for simultaneously migrating multiple virtual machines on a physical machine without exceeding resource limits, leading to potential overload and inefficient migration times.
Innovation Solution
Determine a sequence for live migration of virtual machines based on their migration costs, prioritizing those with lower costs first, and utilizing idle resources after initial migrations to efficiently allocate computing resources for subsequent migrations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple virtual machines are migrated simultaneously, then migration throughput is improved, but computing resource overload occurs
Solution Approach 1:
The patent segments the migration process into multiple phases: initial migration phase (migrating VMs with lower migration costs first) and subsequent migration phase (migrating remaining VMs after resources are freed). This segmentation allows the system to control resource usage while achieving multi-VM migration, resolving the contradiction between throughput and resource overload.
2Adaptability or versatility
If virtual machines with high migration cost are migrated first, then migration sequence flexibility is improved, but resource consumption increases
Solution Approach 1:
The patent changes the parameter of migration sequence based on migration cost calculations. By calculating migration costs for all VMs and sorting them in ascending order, the system dynamically adjusts the migration sequence to optimize resource consumption while maintaining flexibility in handling different migration scenarios.
3Device complexity
If conventional single VM migration is used, then resource allocation is simple, but migration time is excessive
Solution Approach 1:
The patent performs preliminary actions by calculating migration costs for all VMs before starting migration, sorting them by cost, and identifying which VMs can be migrated in the initial phase. This preliminary planning enables efficient multi-VM migration without excessive complexity, reducing overall migration time while maintaining manageable resource allocation.
Data Source
Figure 1
Figure 2
Figure 3~4a
AI summary
A virtual machine migration method is disclosed. The method includes: obtaining a migration cost of a first VM and a migration cost of a second VM, and performing, based on a case in which the migration cost of the first VM is less than the migration cost of the second VM, live migration on the first VM by using a first computing resource, where the first computing resource is an idle computing resource in a physical machine; and performing, based on a case in which live migration on the first VM is completed, live migration on the second VM by using a second computing resource, where the second computing resource is a computing resource allocated when the first VM runs on the physical machine. In this application, a sequence of performing live migration on VMs is determined based on migration costs of the VMs during migration. When there are few computing resources, a VM with a low migration cost is preferentially live migrated, and a computing resource allocated to the VM on which live migration has been completed becomes idle. Then, a remaining VM with a high migration cost is live migrated by using the idle computing resource. This improves migration efficiency and reduces migration time overheads.