Storage Controller Load Distribution via Migration Indices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage system technologies fail to efficiently perform scale-out and virtual machine migration while considering the state of applications, compliance with laws and rules, and resource usage status of the migration destination, especially when managed by another organization.
Innovation Solution
A storage system comprising multiple computer systems connected via a network, where an administrative server monitors resource usage, calculates indices for cost, credibility, and performance, and determines the best migration destination based on compliance with laws and rules, ensuring secure and efficient scale-out and migration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional load distribution technology is used to migrate virtual machines, then cost and performance are considered, but the state of virtual machines, compliance with laws and rules, and resource usage status of migration destination are not taken into account
Solution Approach 1:
The system pre-calculates and stores indices for each computer system including cost indices, credibility indices (reflecting laws and rules), and performance indices. These indices are prepared in advance before migration decisions are needed, allowing the system to quickly evaluate multiple destinations without complex real-time analysis.
Solution Approach 2:
The patent introduces an intermediary evaluation mechanism that uses predefined indices as mediators between the migration source and potential destinations. The administrative server uses these indices to objectively evaluate and select migration destinations, simplifying the decision process while ensuring compliance with laws and rules through the credibility index.
2Reliability
If manual relocation of applications is performed to comply with laws and rules, then compliance is achieved, but the process is time-consuming and difficult to keep track of ever-changing application states
Solution Approach 1:
The system continuously monitors the state of virtual machines and application workloads, providing feedback to the administrative server. This real-time feedback enables automatic detection of migration needs and triggers migration processes without manual intervention, ensuring compliance while reducing time loss through automated decision-making based on current system state.
Solution Approach 2:
The system enables self-service migration by automatically evaluating migration destinations using pre-calculated indices and executing migrations without manual intervention. The administrative server autonomously monitors application states, determines compliance requirements, selects appropriate destinations, and performs migrations, eliminating the need for manual tracking and relocation.
3Productivity
If resource usage status is collected from scale-out destinations managed by another organization, then resource allocation can be optimized, but security concerns arise
Solution Approach 1:
Instead of directly accessing or collecting detailed resource usage status from external organization's systems, the system uses copied or simulated data representations in the form of indices. The second index from the other computer system uses dimensionless quantities to indicate resource usage status without requiring direct access to the actual system, thus maintaining security while enabling optimization.
Solution Approach 2:
The system transforms detailed resource usage parameters into dimensionless index values that preserve the essential information needed for evaluation without exposing sensitive system details. By changing the parameter representation from detailed status data to aggregated index scores, the system enables resource allocation optimization while maintaining security boundaries between organizations.
Data Source
AI summary
Scale-out of a controller and application migration consider the application status, laws and rules of a scale-out destination, and a resource usage status. If an operation status exceeds a threshold value, an administrative server for a computer system of a storage system calculates a first index for each execution target, including an application, a virtual machine, or a container operating in the computer system, based on information about cost, credibility, or performance when the execution target is migrated to, and caused to operate in, another computer system, and indexing information indicating laws or rules applied to the computer system using indexes. A second index indicating, as a dimensionless quantity, the selection degree of another computer system as a migration destination of the execution target is acquired. A migration destination of the execution target is decided from among other computer systems based on the first index and/or the second index.