Memory System Transaction Atomicity via Spare Region Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data processing systems for portable electronic devices lack efficient methods for storing and managing write data grouped as transactions, particularly in scenarios involving sudden power loss, where data integrity and recovery are compromised.
Innovation Solution
A data processing system that includes a host generating write commands with transaction information, such as transaction ID, commit, and abort information, and a memory system storing write data in normal regions and corresponding transaction information in spare regions, allowing for data validation and recovery during power restoration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If write data is stored in normal regions without separate commit commands, then data storage efficiency is improved, but data integrity during power loss deteriorates
Solution Approach 1:
The memory device is divided into normal regions for data storage and spare regions for transaction information storage. This segmentation allows write data to be stored efficiently in normal regions while transaction commit/abort information is stored separately in spare regions, maintaining data integrity without requiring separate commit commands.
Solution Approach 2:
Transaction information including commit and abort flags is stored in advance in the spare regions corresponding to write data locations. When power is restored after sudden power-off, the memory system can immediately check these pre-stored transaction information to determine whether to rebuild the data, eliminating the need for separate commit commands and ensuring data integrity.
2Reliability
If transaction information is stored in spare regions corresponding to normal regions, then data recovery capability is improved, but memory space utilization deteriorates
Solution Approach 1:
The spare regions, which are traditionally unused or reserved areas in memory devices, are utilized to store transaction information. This multi-functional use of spare regions enables data recovery capability without requiring additional dedicated space, thus maintaining memory space utilization efficiency while improving reliability.
3Device complexity
If separate commit commands are eliminated, then system complexity is reduced, but transaction management capability deteriorates
Solution Approach 1:
The memory system automatically manages transactions by checking the commit/abort information stored in spare regions during power restoration. This self-service mechanism eliminates the need for complex separate commit commands while maintaining transaction management capability, as the system autonomously determines whether to rebuild data based on pre-stored transaction information.
Data Source
AI summary
A data processing system includes a host suitable for generating a plurality of write data grouped into transactions and a plurality of write commands including transaction information of each of the write data; and a memory system suitable for storing the write data in a normal region of a memory device in response to the write commands received from the host, and storing the transaction information included in each of the write commands in a spare region, which corresponds to the normal region, of the memory device.