Variable-Size Journal Scheme for Memory Subsystem Data Integrity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional memory sub-systems face inefficiencies due to the generation and storage of uniform, verbose journal entries, leading to increased space requirements and write amplification, which can result in data loss during asynchronous power loss events and reduce the overall quality of service.
Innovation Solution
A journal scheme that generates and stores compact journal entries of varying sizes, using unique formats for different actions and reducing redundant information, allowing for efficient reconstruction of address mapping data structures during power loss.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If uniform verbose journal entries are generated and stored, then data integrity during power loss is maintained, but space requirements and write amplification increase
Solution Approach 1:
The patent segments journal entries into variable-length formats based on operation type. Different operation types (e.g., program, erase, read) have specialized compact journal entry formats that only include relevant fields, eliminating redundant information while maintaining data integrity for recovery purposes.
Solution Approach 2:
The patent extracts only the essential information needed for each specific operation type into the journal entry. By taking out only the necessary fields for each operation (e.g., only program-specific fields for program operations), the journal entry size is minimized while still providing sufficient information for data recovery during power loss events.
2Loss of information
If uniform verbose journal entries are generated and stored, then all operation details are recorded, but write amplification increases
Solution Approach 1:
The patent divides journal entries into operation-specific segments with variable lengths. Each operation type has a tailored format that records only the necessary details for that operation, reducing the total number of bytes written to storage while preserving all critical operation information needed for recovery.
Solution Approach 2:
The patent changes the parameter of journal entry size from fixed to variable based on operation type. By adjusting the journal entry structure dynamically according to the specific operation being logged, the system minimizes write amplification while ensuring complete operation detail recording through appropriate field selection for each operation category.
3Productivity
If more journal entries are stored per page, then space efficiency improves, but journal entry format complexity increases
Solution Approach 1:
The patent segments the journal entry structure into operation-type-specific formats, where each operation category has its own optimized template. This segmentation allows multiple compact entries per page while managing complexity through structured, predictable formats for each operation type, making parsing and generation more efficient despite the variety of formats.
Solution Approach 2:
The patent creates a universal journaling framework that handles multiple operation types through a common structure with operation-specific variations. The unified approach uses a consistent header and footer format with variable middle sections, allowing the system to store diverse operation details efficiently while maintaining manageable overall structure and simplifying processing.
Data Source
AI summary
An operation performed on a non-volatile memory device is detected. A type of the operation is determined. The type of the operation comprises a type identifier and a type modifier. A journal entry is generated reflecting the operation. The journal entry comprises a body reflecting the operation and a header comprising the type identifier and the type modifier. A combination of the type identifier and the type modifier define a size of the journal entry. The journal entry is written to a volatile memory device.