Disk Drive Translation Table Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current magnetic disk drives face challenges in increasing storage density due to the superparamagnetic effect, limiting them to around 1 Tbit/in2, and require new approaches to improve data density beyond this limit.
Innovation Solution
Implementing shingled writing techniques and a reduced-size translation table using an additional address mapping layer and data encoding schemes like run length encoding to efficiently map logical addresses to physical locations, excluding metadata storage locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If shingled writing techniques are implemented to increase data density, then storage density is improved by a factor of 2.3 or higher, but the complexity of address mapping increases requiring an additional translation layer
Solution Approach 1:
The address mapping is segmented into two distinct layers: an first translation layer that maps host logical addresses to drive logical addresses, and a second translation layer that maps drive logical addresses to physical locations. This segmentation allows each layer to handle specific aspects of the mapping independently, managing the complexity introduced by shingled writing while enabling higher data density through overlapping track writes.
2Reliability
If a full translation table is used to map all logical addresses to physical locations, then complete address mapping is achieved, but the translation table size becomes excessively large consuming significant memory resources
Solution Approach 1:
The translation table functionality is divided into two separate translation layers, each maintaining a smaller table for its specific mapping scope. The first translation layer maintains a table mapping host logical addresses to drive logical addresses, while the second layer maintains a table mapping drive logical addresses to physical locations. This segmentation reduces the size of individual translation tables while maintaining complete address mapping capability.
Solution Approach 2:
The patent introduces an additional dimension to the address mapping by inserting a drive logical address layer between host logical addresses and physical locations. This dimensional expansion creates a hierarchical mapping structure where data can be accessed through multiple address representations, reducing the need for a single large translation table while maintaining complete mapping capability.
3Ease of operation
If metadata is stored at every physical location to enable random access, then random read access is enabled, but the usable data storage capacity is reduced due to metadata overhead
Solution Approach 1:
Metadata is segmented and stored selectively at specific physical locations rather than at every location. The system identifies and stores metadata only where necessary to enable random read access, reducing the overall metadata overhead and increasing usable data storage capacity while maintaining the required access capabilities.
Solution Approach 2:
The patent extracts the metadata storage requirement from every physical location and consolidates it to only those locations where it is truly necessary for random access. By taking out the metadata from unnecessary locations, the system reduces storage overhead while preserving random read access functionality.
Data Source
AI summary
A disk drive is disclosed that utilizes an additional address mapping layer between logical addresses used by a host system and physical locations in the disk drive. Physical locations configured to store metadata information can be excluded from the additional address mapping layer. As a result, a reduced size translation table can be maintained by the disk drive. Improved performance, reduced costs, and improved security can thereby be attained.