Storage Device Section Management for Data Access Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage devices struggle to efficiently manage data access patterns and divide logical address spaces based on spatial locality and historical request information, leading to inflexible data management and difficulty in discriminating associated data.
Innovation Solution
A data managing method and storage device with a controller that generates sections based on input/output requests, using spatial locality and historical request information to manage logical addresses, allowing for flexible division of address spaces and efficient data management by merging or dividing sections as needed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional storage devices manage data access patterns without spatial locality and historical request information, then the device complexity is low, but the data access efficiency is poor and the ability to discriminate associated data is limited
Solution Approach 1:
The logical address space is divided into multiple sections based on spatial locality information. Each section contains a range of logical addresses that are accessed together, allowing the storage device to manage and access data more efficiently by operating on sections rather than individual addresses.
Solution Approach 2:
The section boundaries and spatial locality information are dynamically updated based on historical request information. The storage device learns from past access patterns and adjusts the section divisions to match actual data access behavior, improving efficiency over time while adapting to changing workloads.
2Adaptability or versatility
If the storage device uses fixed address space management, then the device complexity is low, but the adaptability to different data access patterns is poor
Solution Approach 1:
The storage device dynamically adjusts section boundaries based on historical request information, allowing the address space management to adapt to different data access patterns. The system learns from past access behavior and reorganizes sections to match actual usage patterns, providing versatility without requiring manual configuration.
Solution Approach 2:
The storage device uses historical request information as feedback to continuously improve its section management. By analyzing past access patterns and using this information to adjust future section divisions, the system adapts to varying data access patterns automatically.
3Productivity
If the storage device divides logical address spaces into sections based on spatial locality and historical requests, then the data access efficiency improves, but the difficulty of managing and discriminating data increases
Solution Approach 1:
By dividing the logical address space into clearly defined sections with specific boundary information, the storage device makes data discrimination easier. Each section is associated with specific spatial locality information, allowing the system to quickly identify and manage relevant data ranges without confusion.
Solution Approach 2:
The storage device uses spatial locality information as a distinguishing characteristic for each section, similar to using different colors to mark different groups. This metadata acts as an identifier that makes it easy to detect and measure which section a logical address belongs to, simplifying data discrimination.
4Manufacturing precision
If the storage device uses detailed section information including historical request numbers, then the data management precision improves, but the loss of information storage capacity increases
Solution Approach 1:
The storage device stores only essential historical request information (such as request counts and basic access patterns) rather than complete detailed histories. This partial information is sufficient to achieve effective section management and data discrimination while minimizing the overhead on storage capacity.
Data Source
AI summary
Disclosed is a data managing method of a storage device which includes at least one nonvolatile memory device and a controller controlling the nonvolatile memory device. The data managing method includes receiving an input/output request and generating a section directing logical addresses based on the input/output request. The section is managed using section information, and the section information includes a start logical address corresponding to the input/output request, spatial locality information having the number of the directed logical addresses, and historical request information.


