Automated Data Migration Scheduling via Storage Load Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data migrations in data centers are manually planned, which can lead to inefficiencies due to varying workloads and loading configurations of source and target storage systems, making it challenging to identify optimal time windows for non-disruptive data transfers.
Innovation Solution
A method and apparatus that calculate source and target storage system utilization scores, compute a source-target load score for each time window, and select the most suitable window or set of windows for data migration based on these scores to automate the data migration process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data migration is performed manually without automated scheduling, then flexibility in planning is maintained, but migration efficiency decreases and disruptions increase due to inability to identify optimal time windows
Solution Approach 1:
The storage system automatically monitors its own workload conditions and schedules migrations without external intervention. The system calculates workload metrics, identifies optimal time windows, and executes migrations autonomously, eliminating the need for manual scheduling while maintaining operational simplicity
Solution Approach 2:
The system dynamically adjusts migration scheduling based on changing workload parameters. By continuously monitoring workload metrics and adapting migration timing to current system conditions, the system optimizes migration efficiency without requiring complex manual intervention or rigid scheduling protocols
2Speed
If data migration is scheduled during high-utilization periods, then migration speed may be maintained, but system performance and reliability deteriorate due to resource contention
Solution Approach 1:
The migration scheduling system dynamically adapts to changing workload conditions by continuously monitoring utilization metrics and adjusting migration timing. Rather than fixed scheduling, the system flexibly selects time windows based on real-time system state, ensuring migrations occur during low-utilization periods when system stability is maintained
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring workload metrics during migration operations. This feedback loop allows the system to detect when resource contention begins to impact system stability and adjust or pause migrations accordingly, maintaining both migration progress and system reliability
3Ease of operation
If data migration is performed without workload analysis, then operational simplicity is maintained, but migration disruptions increase and service quality deteriorates
Solution Approach 1:
The storage system automatically performs workload analysis and schedules migrations without requiring manual intervention. The system independently monitors its own performance metrics, identifies optimal migration windows, and executes migrations, eliminating operational complexity while minimizing disruptions through intelligent scheduling
Solution Approach 2:
The system monitors and responds to changing workload parameters by adjusting migration timing accordingly. By detecting patterns in system utilization and adapting migration schedules to low-utilization periods, the system reduces migration disruptions while maintaining operational simplicity through automated decision-making
Data Source
AI summary
Data migration between a source storage system and a target storage system is automated by calculating the optimal time window or group of consecutive time windows in which to execute the data migration, where optimality is defined in terms of source storage system and target storage system utilizations combined in a manner that accounts for differences between the source storage system and target storage system utilizations. Source and target storage system utilization scores are calculated for each of a plurality of time windows of at least one representative time period. A source-target load score is calculated for each of the time windows based on the source storage system utilization scores and the target storage system utilization scores. At least one of the time windows is selected based on the source-target load scores.