VM Migration Destination Selection Using Resource Usage Pattern Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
When migrating virtual machines between buildings, it is challenging to determine a suitable migration destination that avoids resource depletion on the physical servers at the destination, especially during large-scale power outages, due to the complexity of resource conflicts and varying resource demands.
Innovation Solution
A migration destination determination apparatus that extracts resource usage patterns from multiple physical servers and determines a suitable destination based on the similarity between the resource usage patterns of the virtual machine and the physical servers, using a system that includes an extraction unit for pattern extraction and a migration destination determination unit to evaluate potential destinations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If virtual machines are migrated between buildings during power outages, then service continuity is maintained, but resource conflicts and depletion occur at migration destinations
Solution Approach 1:
The system performs preliminary analysis of resource usage patterns and predicts future resource requirements before migration occurs. By extracting variation patterns from historical data and projecting them forward, the system identifies suitable migration destinations in advance, ensuring that resource conflicts are avoided before they occur during actual migration operations.
Solution Approach 2:
The system continuously monitors resource usage at both source and destination buildings, comparing actual consumption against predicted patterns. This feedback mechanism allows the system to validate migration decisions and adjust future migration planning based on observed resource behavior, preventing resource depletion at destination sites.
2Reliability
If comprehensive resource confirmation is performed for all migration source-destination combinations, then resource depletion is avoided, but the complexity and time required for determination increases enormously
Solution Approach 1:
The system transforms the complex multi-dimensional migration determination problem into a simpler form by changing parameters: instead of evaluating all possible source-destination combinations, it extracts key variation patterns from resource usage data and uses these patterns to rapidly assess suitability. This parameter transformation reduces computational complexity while maintaining reliable resource conflict detection.
Solution Approach 2:
The system extracts essential variation patterns from comprehensive resource usage data, separating the critical information needed for migration decisions from the overwhelming volume of raw data. By taking out only the relevant variation patterns rather than processing all data points, the system achieves reliable resource conflict avoidance with reduced computational complexity.
3Productivity
If migration destinations are selected without considering resource usage patterns, then migration speed is increased, but resource depletion and conflicts occur at destinations
Solution Approach 1:
The system performs preliminary extraction and analysis of resource usage variation patterns before migration decisions are made. By having this pattern analysis prepared in advance, the system can rapidly determine suitable migration destinations without sacrificing resource availability, achieving both high migration speed and reliable resource management.
Data Source
AI summary
A migration destination determination apparatus configured to, in migration of a virtual machine, determine a physical server as a migration destination of the virtual machine, the migration destination determination apparatus including: an extraction unit configured to extract one or more variation patterns in resource usage from resource amount logs acquired from a plurality of physical servers; and a migration destination determination unit configured to determine, based on a similarity between a variation pattern in resource usage of the virtual machine in an area of a migration source and a variation pattern in resource usage of a physical server in another area, the physical server as a migration destination of the virtual machine.