Digital Twin Storage Tier Mapping for Accurate Object Placement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage tiering decisions are often based on intuition and limited factors, leading to inefficient allocation of storage objects to storage tiers, which can result in suboptimal use of resources and increased costs.
Innovation Solution
Employing digital twins of storage tiers to evaluate storage object placement scores using telemetry data for various placement options, determining the best placement based on designated storage metrics and telemetry data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If storage objects are assigned to storage tiers based on intuition and limited factors, then the storage system is easy to operate, but the allocation efficiency and resource utilization deteriorate
Solution Approach 1:
The patent creates a digital twin - a virtual copy of the storage system that replicates its structure, metrics, and behavior. This digital twin is used to simulate and evaluate different storage tier assignment scenarios without affecting the actual storage system operations, enabling efficient optimization through virtual modeling
Solution Approach 2:
The patent replaces manual, intuitive storage tier assignment with an automated system that uses machine learning models and digital twins to determine optimal placements. The system automatically evaluates multiple scenarios and selects the best assignment based on predicted performance metrics
2Device complexity
If storage objects are assigned to storage tiers based on limited factors, then the device complexity is low, but the measurement precision of placement effectiveness deteriorates
Solution Approach 1:
The digital twin continuously incorporates feedback from actual storage system performance data, including metrics like access patterns, throughput, and cost information. This feedback loop enables the system to refine its predictions and improve the precision of placement effectiveness measurements over time
Solution Approach 2:
The system evaluates multiple storage tier assignment scenarios by changing and comparing different parameter combinations (e.g., different allocation strategies, tier configurations, and weightings). This allows precise measurement of placement effectiveness across various conditions without increasing actual system complexity
3Ease of manufacture
If conventional storage tiering is used, then the implementation is simple, but the cost efficiency and resource utilization deteriorate
Solution Approach 1:
The system performs preliminary actions by creating and configuring the digital twin before making actual storage tier assignment decisions. It pre-evaluates multiple scenarios and identifies optimal placements in advance, allowing for cost-efficient resource allocation without complicating the actual implementation
Data Source
AI summary
Techniques are provided for mapping storage objects to storage tiers using digital twins. One method comprises initializing at least one virtual representation for multiple storage tiers of a storage system using storage metrics of the respective storage tier, wherein the storage system comprises storage objects, and wherein a storage object is mapped to a particular storage tier according to a storage object to storage tier placement option; configuring the at least one virtual representation, for a plurality of iterations, based on telemetry data for the storage objects, wherein each iteration corresponds to a different storage object to storage tier placement option and generates a storage object placement score for the respective storage object to storage tier placement option; selecting a given storage object to storage tier placement option based on the respective storage object placement score; and initiating an implementation of the selected storage object to storage tier placement option.