Dynamic Storage Tiering Orchestrator for Data Access Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing systems face challenges in efficiently managing client data storage due to limited resources and the need for seamless transitions between different storage types, which can lead to data loss and performance issues.
Innovation Solution
A data storage system that includes a data storage orchestrator to dynamically assign high or low performance resources based on client usage rates, providing data redundancy and seamless transitions between storage types to ensure continuous service and optimal resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is transferred to tiered storage system, then storage capacity is improved, but data access speed deteriorates
Solution Approach 1:
The patent implements dynamic tiering that automatically moves data between high-speed and capacity storage tiers based on access patterns. Hot data frequently accessed by clients is dynamically transferred to high-speed storage, while cold data is moved to capacity storage, optimizing both access speed and storage utilization without manual intervention.
Solution Approach 2:
The patent uses a data flow analogy similar to fluid dynamics, where data moves between storage tiers based on pressure (access demand). High-access data experiences 'pressure' that pushes it to high-speed storage, while low-access data naturally settles in capacity storage, creating an automated balancing system.
2Speed
If storage resources are allocated to high performance, then data access speed is improved, but resource cost deteriorates
Solution Approach 1:
The patent applies local quality by providing high-performance storage resources selectively to specific data sets based on their access patterns. Instead of uniformly allocating high-performance resources to all data, the system identifies hot data and concentrates performance resources only where needed, reducing overall resource costs while maintaining access speed for critical data.
Solution Approach 2:
The system dynamically changes storage performance parameters by adjusting the performance tier assignment of data sets based on monitored access patterns. When access frequency exceeds thresholds, the system transitions data from capacity storage to high-speed storage, effectively changing the performance parameter to match actual usage requirements.
3Productivity
If data is moved between storage tiers, then storage efficiency is improved, but data loss risk increases
Solution Approach 1:
The patent implements beforehand cushioning by maintaining redundant data copies during the tiering transition process. Before moving data from one tier to another, the system creates backup copies and verifies data integrity, ensuring that no data is lost during the transfer process. This cushioning mechanism protects against potential data loss from transfer failures.
4Measurement precision
If manual data management is performed, then control precision is improved, but operational complexity deteriorates
Solution Approach 1:
The patent implements self-service by enabling the storage system to automatically monitor its own data access patterns and perform tiering decisions without external intervention. The system autonomously identifies hot and cold data, makes placement decisions, and executes data movement, eliminating the need for manual data management while maintaining precise control over data placement.
Solution Approach 2:
The system employs feedback mechanisms where access pattern data is continuously collected, analyzed, and used to adjust data placement decisions. The tiering system receives feedback from client access behavior and automatically adjusts data tier assignments, creating a closed-loop control system that achieves precise control without manual intervention.
Data Source
Figure 1.1
Figure 1.2
Figure 1.3
AI summary
A data storage system for managing storage of data from clients includes a data storage and a data storage orchestrator. The data storage includes an accelerator pool and a non-accelerator pool. The data storage orchestrator identifies a client assignment update event based on a change in use of the data storage by a client of the clients, makes a determination that the client assignment update event is a promotion event, and in response to the determination: promotes the client to move a primary data storage of the client from the non-accelerator pool to the accelerator pool.