Switch-Based Virtual Machine Migration for Energy Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtual machine migration technologies in data centers often lead to packet loss due to improper migration, as the virtual machine management server lacks real-time usage condition data, resulting in inefficient energy saving and service disruption.
Innovation Solution
A method where a switch detects low virtual machine usage by monitoring port traffic and requests the virtual machine management server to migrate virtual machines to other servers, ensuring migrations are based on actual usage conditions, thereby reducing packet loss and energy consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If virtual machine migration is performed based on server resource utilization, then energy consumption is reduced, but packet loss occurs due to lack of real-time usage condition data
Solution Approach 1:
The patent implements feedback by having the network device monitor port traffic conditions and send this real-time usage information to the virtual machine management server. The management server then uses this feedback to make informed migration decisions, ensuring that migrations occur only when appropriate (e.g., when traffic is low), thus reducing packet loss while achieving energy savings.
Solution Approach 2:
The patent applies preliminary action by having the network device continuously monitor and report traffic conditions before migration decisions are made. This advance information allows the management server to plan migrations in advance during low-traffic periods, preventing packet loss while still achieving the energy savings from consolidation.
2Use of energy by stationary object
If virtual machine migration is performed to consolidate resources, then energy consumption is reduced, but service disruption occurs due to improper migration timing
Solution Approach 1:
The system uses feedback from continuous traffic monitoring to determine optimal migration timing. The network device reports real-time traffic conditions to the management server, which schedules migrations during periods of low activity, thus maintaining service continuity while achieving energy savings through resource consolidation.
Solution Approach 2:
The patent makes the migration policy dynamic by adjusting migration decisions based on real-time traffic conditions. Rather than using fixed schedules or static resource thresholds, the system adapts migration timing to actual network usage patterns, ensuring service continuity while enabling energy savings.
3Device complexity
If virtual machine migration is performed without real-time usage data, then migration overhead is reduced, but migration accuracy decreases leading to packet loss
Solution Approach 1:
The patent extracts the traffic monitoring and reporting function from the management server and places it at the network device level. This allows local collection of precise usage data without adding significant complexity to the overall system, as the network device is already equipped with traffic monitoring capabilities that are now being utilized for migration decision-making.
Data Source
Figure 1~2
Figure 3~5
Figure 6~7
AI summary
A method, an apparatus and a system for migrating virtual machines are provided. The method includes: a switch detecting whether a quantity of the virtual machines connected with a first virtual machine bearing server is less than a first threshold; when the quantity of the virtual machines is less than the first threshold, the switch requesting a virtual machine management server to migrate the virtual machines connected with the first virtual machine bearing server to another virtual machine bearing server or other virtual machine bearing servers. The embodiments of the present invention solve the problem that packet loss is possibly caused by virtual machine migration methods in relevant technologies, and guarantee the smooth progress of services while reducing the energy consumption of data centre network devices.