Mapping Storage Objects with Digital Twins for Load Balancing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage object placement techniques fail to optimize resource utilization of storage controllers due to varying IO activity over time, leading to imbalanced load distribution and inefficient resource management.
Innovation Solution
Employ digital twins to simulate storage object to storage controller mappings, evaluating load balance scores based on storage metrics to dynamically adjust placements and optimize resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If storage objects are assigned to storage controllers based on initial mapping policy, then storage objects can be deployed and processed, but load distribution becomes imbalanced over time due to varying IO activity
Solution Approach 1:
The patent implements dynamic remapping of storage objects to storage controllers based on real-time load metrics. The system continuously monitors IO activity and automatically adjusts mappings to balance load distribution, transforming a static mapping policy into a dynamic adaptive system that responds to changing workload conditions.
Solution Approach 2:
The system employs feedback mechanisms by monitoring storage controller load metrics and using this information to adjust storage object mappings. The feedback loop ensures that mappings are optimized based on actual performance data, enabling continuous improvement of load distribution and resource utilization.
2Ease of operation
If manual storage object placement is used, then placement decisions can be controlled, but system complexity increases and automation is reduced
Solution Approach 1:
The system performs self-optimization by automatically analyzing load metrics and adjusting storage object mappings without requiring manual intervention. The automated remapping mechanism reduces operational complexity while maintaining control through policy-based decision-making, allowing the system to manage itself based on predefined optimization criteria.
3Stability of the object's composition
If storage object mappings are fixed, then system stability is maintained, but adaptability to changing IO patterns is lost
Solution Approach 1:
The patent transitions from fixed static mappings to dynamic adaptive mappings that automatically adjust based on monitored IO patterns and load conditions. This dynamic approach enables the system to adapt to changing workload characteristics while maintaining stability through automated, policy-driven decision-making processes.
Data Source
AI summary
Techniques are provided for mapping storage objects to storage controllers using digital twins. One method comprises obtaining a virtual representation of a storage system that comprises storage objects and storage controllers, wherein a given storage object is mapped to a particular storage controller according to a storage object to storage controller mapping configuration; configuring the virtual representation of the storage system, for multiple iterations, based on at least one storage metric for respective storage objects, wherein each iteration corresponds to a different storage object to storage controller mapping configuration and generates a load balance score for the respective storage object to storage controller mapping configuration; selecting a given storage object to storage controller mapping configuration based on the respective load balance scores; and initiating an implementation of the selected storage object to storage controller mapping configuration in the storage system.