VM Consolidation via Parallel Migration and Interleaved Packing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Data centers face high power consumption costs due to the need for multiple physical machines (PMs) to support virtual machines (VMs), with existing power management techniques not effectively reducing energy usage, especially for components like memories and hard disks.
Innovation Solution
A method and manager PM for VM consolidation that classifies PMs as redundant and surviving, using an interleaved N-dimensional best-fit decreasing packing algorithm to migrate VMs to minimize the number of PMs required and optimize resource utilization, thereby maximizing the number of PMs that can be turned off, while minimizing migration time and ensuring load balancing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by stationary object
If VM consolidation is performed to reduce the number of PMs, then power consumption is reduced, but VM migration time increases
Solution Approach 1:
The patent implements dynamic parallel migration where multiple VMs are migrated simultaneously across multiple networks instead of sequentially. The migration manager coordinates parallel migration tasks, allowing concurrent data transfer between source and destination PMs, thereby reducing total migration time while achieving consolidation goals
Solution Approach 2:
The patent transitions from one-dimensional sequential migration to multi-dimensional parallel migration by utilizing multiple networks for simultaneous VM migration. This dimensional expansion allows multiple migration operations to occur concurrently, resolving the time penalty associated with consolidation
2Loss of time
If multiple VMs are migrated in parallel, then migration time is reduced, but network resource consumption increases
Solution Approach 1:
The patent applies partial action by selectively migrating only necessary VMs to target PMs based on consolidation requirements, rather than migrating all VMs. The migration manager intelligently selects which VMs to move, avoiding unnecessary network resource consumption while still achieving the consolidation objective
Solution Approach 2:
The migration manager acts as an intermediary that coordinates parallel migrations across multiple networks. It manages and balances network resource allocation, ensuring that parallel migration tasks do not overload any single network, thereby reducing overall network resource consumption while maintaining migration speed
3Quantity of substance
If VMs are migrated to minimize the number of PMs, then the number of redundant PMs increases, but VM migration complexity increases
Solution Approach 1:
The patent segments the migration process into manageable tasks coordinated by a migration manager. The complex consolidation problem is divided into individual VM migration tasks that can be planned and executed separately, reducing overall complexity. Each VM migration is treated as an independent task that can be managed through standardized procedures
Solution Approach 2:
The migration manager serves as an intermediary that simplifies the complexity of VM consolidation. It automatically generates migration plans, selects target PMs, and coordinates execution, thereby reducing the operational complexity despite increasing the number of redundant PMs available for migration
Data Source
AI summary
A method and a manager physical machine (PM) for virtual machine (VM) consolidation are provided. The method is performed by the manager PM. A network connects the manager PM and a plurality of server PMs. A plurality of VMs is running on the server PMs. The method includes the following steps. The manager PM classifies the server PMs into redundant PMs and surviving PMs. The manager PM determines migration paths of the VMs running on the redundant PMs to the surviving PMs. The manager PM determines a parallel migration sequence of the VMs running on the redundant PMs based on the migration paths. The manager PM migrates the VMs running on the redundant PMs to the surviving PMs in parallel according to the parallel migration sequence.