Distilled Protection Sector for Missed Write Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage systems, such as HDDs and SSDs, face issues with data integrity due to errors like 'missed writes' where data is reported as written but not recorded, which can be caused by drive failures or power loss, and these systems lack effective mechanisms to detect and correct such errors.
Innovation Solution
Implementing a data storage device that utilizes the T10 Protection Information Model (PIM) by copying protection information to a distilled protection sector and maintaining version information in a separate database, allowing for detection of missed writes and other data storage errors by verifying the version information before data is read.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If protection information is copied to a distilled protection sector and version information is tracked in a separate database, then data integrity and detection of missed writes is improved, but device complexity and storage space requirements increase
Solution Approach 1:
The protection mechanism is segmented into multiple independent components: protection information stored with data sectors, a distilled protection sector containing condensed protection data, and a separate database for version tracking. This segmentation allows each component to serve a specific function while collectively providing robust data integrity verification without requiring a complete redesign of the storage system.
Solution Approach 2:
Protection information is copied from data sectors to a distilled protection sector, creating a redundant verification mechanism. The copying process preserves essential protection data in a condensed format that can be independently verified against the original data sectors, enabling detection of missed writes without storing full redundancy of all protection information.
2Reliability
If protection information bytes are stored with data in each sector and copied to distilled protection sector, then detection capability for missed writes is improved, but storage space is consumed
Solution Approach 1:
Essential protection information is extracted from full data sectors and stored in a distilled protection sector. This extraction process identifies and preserves only the critical protection bytes needed for verification, separating them from the main data storage areas. The distilled sector serves as a compact verification repository that reduces the need for extensive protection information storage across all data sectors.
Solution Approach 2:
Instead of storing complete protection information redundantly in every sector, the system applies partial action by storing full protection information only where necessary (with data sectors) and using a distilled version for verification. This approach provides sufficient protection capability while avoiding excessive storage consumption that would result from full redundancy throughout the entire storage system.
Data Source
AI summary
A write command includes data targeted to a data sector of a non-volatile memory. The data is committed to be stored in the data sector. Protection information bytes of the data are stored together with the data in the data sector, and the protection information bytes are copied to a distilled protection sector. A read request is received for the data and in response, the protection information in the data sector is validated against the copy of the protection information in the distilled protection sector.