Storage Migration Candidate Ranking via Performance Impact Scores

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Determining when and how to redistribute stored data in data centers to avoid performance problems is challenging due to the difficulty in assessing the relationship between storage and performance capacity.

Innovation Solution

A method and apparatus that calculate a performance impact score for each application, representing the relationship between current and maximum utilized storage and performance capacity, to prompt data migration actions based on these scores.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If the storage system increases the number of host servers and storage capacity to accommodate increased demand, then the storage capacity and computing capabilities are improved, but the difficulty in determining when and how to redistribute stored data increases

Engineering Contradiction:

Improvestorage capacityVSAvoiddifficulty in determining data redistribution

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The system continuously monitors storage capacity utilization and performance metrics, using this feedback to automatically calculate performance impact scores and determine optimal migration timing. The feedback loop enables the system to adapt to changing workload patterns and make informed redistribution decisions without manual intervention.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The storage system performs self-assessment by calculating performance impact scores for each application based on its current storage capacity utilization and projected future utilization. This self-service capability allows the system to autonomously identify migration candidates and determine optimal redistribution strategies without external manual analysis.

Inventive Principle:

Principle #25Self-service

2Reliability

If the storage system monitors and manages data redistribution, then performance problems are avoided, but the complexity of managing storage and performance capacity increases

Engineering Contradiction:

Improveperformance problem avoidanceVSAvoidcomplexity of managing storage capacity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system uses continuous monitoring of storage capacity utilization and performance metrics as feedback to automatically adjust data redistribution strategies. This feedback mechanism enables proactive performance management by identifying potential issues before they occur and implementing corrective actions automatically.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts management parameters such as migration thresholds, priority levels, and redistribution timing based on real-time performance data. By changing these parameters adaptively rather than using fixed rules, the system manages complexity while maintaining high reliability.

Inventive Principle:

Principle #35Parameter changes

3Ease of operation

If the system calculates performance impact scores for all applications, then informed decision-making for data redistribution is enabled, but the computational overhead increases

Engineering Contradiction:

Improveinformed decision-makingVSAvoidcomputational overhead

Core Design Contradiction:

Ease of operationVSUse of energy by moving object

Solution Approach 1:

The system calculates performance impact scores for all applications to ensure comprehensive decision-making, but optimizes the calculation process by using sampled performance data and incremental updates rather than complete reanalysis. This partial action approach maintains decision quality while reducing computational overhead.

Inventive Principle:

Principle #16Partial or excessive action

Solution Approach 2:

The storage system performs self-assessment by calculating performance impact scores autonomously based on its own monitored metrics. This self-service capability eliminates the need for external analysis tools and reduces overall system computational overhead by leveraging the storage system's existing monitoring resources.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11847351B2Automatic identification and ranking of migration candidate storage groups based on relative performance impact to current storage array components

Publication Date: 2023.12.19 DELL PROD LP

AI summary

Storage object groups uniquely associated with respective host applications are processed to model, for each host application, the relationship between current utilized storage capacity of each host application and greatest possible utilized storage capacity of each host application without exhausting either the storage capacity of the storage system or the performance capacity of the storage system. The modeled relationships may be used to calculate headroom and performance impact scores for each host application. Storage object groups that have insufficient headroom for growth, e.g., as indicated by performance impact score, are deemed to be associated with host application workloads that are candidates for migration to a different storage system. The candidates may be ranked and selected for migration based on performance impact scores.