Memory ECC History Table for Write Endurance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As memory density increases and bit error rates rise, memory devices face write endurance issues and power dissipation challenges due to frequent re-write operations, especially in flash and phase change memory devices, where numerous write operations can lead to reduced lifespan and increased power consumption.
Innovation Solution
A memory system and method that involves accessing preexisting data, identifying and correcting errors, and storing error information without overwriting the original data, thereby minimizing write operations and reducing power dissipation by using an error history table to track and correct errors without rewriting the entire page.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If error correction is performed by rewriting corrected data back into the memory array, then data integrity is maintained, but write operations increase causing reduced lifespan and increased power consumption
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing corrected data in a buffer before actual read operations occur. When errors are detected during reading, the corrected data is already prepared in the buffer, eliminating the need to rewrite corrected data back into the memory array and thus extending memory lifespan.
Solution Approach 2:
The patent uses copying by creating a buffer that stores copies of corrected data. Instead of rewriting corrected data back into the original memory array, the system copies the corrected data into the buffer and uses this copy for subsequent read operations, reducing write operations to the memory array and extending its lifespan.
2Reliability
If corrected data is rewritten back into the memory array, then data accuracy is improved, but power dissipation increases
Solution Approach 1:
The system performs preliminary correction of data and stores corrected versions in a buffer before actual read operations. This preliminary action eliminates the need for subsequent rewrite operations, significantly reducing power dissipation while maintaining data accuracy.
Solution Approach 2:
The patent implements copying by storing corrected data in a buffer rather than rewriting it back into the memory array. This copying approach maintains data accuracy while avoiding the power-consuming rewrite operations, thus reducing overall power dissipation.
3Reliability
If numerous rewrite operations are performed to correct errors, then data reliability is maintained, but write endurance is reduced
Solution Approach 1:
The patent applies preliminary action by pre-correcting data and storing it in a buffer before read operations. This eliminates the need for numerous rewrite operations to correct errors, thereby maintaining data reliability while preserving write endurance by reducing the number of writes to the memory array.
Solution Approach 2:
The system uses copying by storing corrected data in a buffer instead of rewriting it back into the memory array. This copying mechanism maintains data reliability through error correction while preserving write endurance by minimizing write operations to the array.
4Reliability
If the entire page is rewritten to correct errors, then data integrity is restored, but time consumption increases
Solution Approach 1:
The patent applies preliminary action by pre-computing corrected data in a buffer before actual read operations. When errors are detected, the corrected data is already prepared, eliminating the need to rewrite the entire page and significantly reducing correction time while maintaining data integrity.
Solution Approach 2:
The system uses copying by storing corrected data in a buffer rather than rewriting the entire page back into the memory array. This copying approach restores data integrity through error correction while reducing time consumption by avoiding full page rewrites.
Data Source
AI summary
A method of reading data from a memory device is disclosed. The method comprises accessing preexisting data from a location in the memory device in response to a read command and identifying an error in the preexisting data. The identified error is corrected for transmission as corrected data. Error information representing the identified error is stored while the preexisting data is retained in the location of the memory device in uncorrected form.