Tiered Storage Controller Managing Data Migration via Heat Maps

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current tiered storage systems face challenges in optimizing performance, reducing complexity, and minimizing wear on solid state memories, particularly due to the increasing cost of high-performance storage and the need for redundancy, while balancing the benefits of different storage mediums like SSDs and HDDs.

Innovation Solution

A tiered storage system architecture that employs a first tier with high-performance storage for frequently accessed data and a second tier for less frequently accessed data, using a tier map controller to manage data migration based on access history and maintain a heat map to determine 'hot' and 'cold' data, thereby reducing write amplification and storage media wear.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If data is stored on high-performance storage tiers, then access speed is improved, but storage cost increases

Engineering Contradiction:

Improvedata access speedVSAvoidstorage capacity

Core Design Contradiction:

SpeedVSQuantity of substance

Solution Approach 1:

The storage system is divided into multiple tiers (first storage tier with fast storage media, second storage tier with slower but cheaper storage media). Data is segmented and placed in different tiers based on access patterns, allowing the system to achieve both high performance for frequently accessed data and cost-effectiveness for less frequently accessed data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Different storage tiers are assigned different quality characteristics (speed, cost) based on local data access requirements. The first tier provides high-speed access for hot data, while the second tier provides cost-effective storage for cold data, creating a non-uniform quality distribution that optimizes both performance and cost.

Inventive Principle:

Principle #3Local quality

2Ease of operation

If manual data tiering is implemented, then data organization is improved, but system complexity increases

Engineering Contradiction:

Improvedata organizationVSAvoidsystem complexity

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The system automatically performs data tiering based on access history and heat maps without requiring manual intervention. The controller monitors data access patterns, updates heat maps, and autonomously migrates data between tiers, eliminating the need for complex manual data organization processes.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system continuously monitors data access patterns and uses this feedback to update heat maps and adjust data tiering decisions. This closed-loop feedback mechanism enables the system to adapt to changing data access patterns automatically, reducing the need for complex manual reorganization.

Inventive Principle:

Principle #23Feedback

3Productivity

If frequent write operations are performed on solid state memory, then data updates are improved, but storage media wear increases

Engineering Contradiction:

Improvedata update speedVSAvoidstorage media lifespan

Core Design Contradiction:

ProductivityVSDuration of action of stationary object

Solution Approach 1:

The system performs preliminary actions by migrating infrequently accessed data to the second storage tier before it accumulates excessive write operations on solid state memory. This preventive measure reduces write amplification on the first tier's solid state media, extending its lifespan while maintaining high update performance for frequently accessed data.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system uses the second storage tier (with cheaper, more durable storage media) to store data that would otherwise cause excessive wear on the expensive solid state memory. This allows the solid state memory to be used only for data that requires frequent updates, effectively extending its operational lifespan.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

4Reliability

If redundancy is added to storage systems, then data reliability is improved, but system complexity and cost increase

Engineering Contradiction:

Improvedata reliabilityVSAvoidsystem complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system segments data into different tiers based on access patterns and reliability requirements, rather than implementing redundancy across the entire storage system. This allows the system to achieve adequate reliability through intelligent data placement while avoiding the complexity and cost of full redundancy mechanisms.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS10176212B1Top level tier management

Publication Date: 2019.01.08 SEAGATE TECH LLC

AI summary

Systems and methods are disclosed for management of a tiered storage system by a top tier storage device. In some embodiments, an apparatus may comprise a circuit configured to maintain an address map at a first storage tier, receive a read request for specified data, return the specified data when the data exists on the first storage tier, and when the specified data does not exist on the first storage tier, return an indication to query a second storage tier. The circuit may be further configured to determine infrequently accessed cold data stored to the first tier, provide to a host device a copy of the cold data stored in an area of the first storage tier scheduled for defragmentation, and perform the defragmentation operation, including copying valid data to an available area of the first storage tier, the valid data not including the cold data.