Self-Identifying Memory Faults in Phase Change Cells
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory devices face issues with memory wear-out, permanent errors causing 'stuck-at' bits, and transient errors due to faulty cells, which existing error detection and correction schemes struggle to address effectively, especially when fault location information is lost during memory failures.
Innovation Solution
Implementing durable self-identifying faults in memory cells that can persistently indicate faulty states, allowing for replacement encoding techniques to steer around faulty cells, using write-verify processes and replacement cells to manage and correct errors at various granularities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional error detection and correction schemes are used, then soft errors can be corrected, but hard errors cause loss of fault location information when storing in another memory location
Solution Approach 1:
The memory cell is modified to self-identify its faulty state through an additional state that automatically indicates when the cell cannot be programmed or has failed, eliminating the need for external fault tracking mechanisms
Solution Approach 2:
The memory cell is prepared in advance with an additional state that can be set to indicate a faulty condition before actual data storage operations occur, allowing proactive fault identification rather than reactive tracking
2Reliability
If memory cells are replaced one at a time or in clusters, then faulty cells can be identified and replaced, but storage overhead and complexity increase
Solution Approach 1:
The fault identification capability is implemented locally within each individual memory cell rather than requiring global tracking mechanisms, allowing each cell to independently indicate its own faulty state through its additional state
3Measurement precision
If fault location information is recorded during write operations, then faulty cells can be identified, but the information is lost if the storage location fails
Solution Approach 1:
Each memory cell independently tracks and indicates its own faulty state through an additional state, making the fault information self-contained within the cell rather than relying on external storage that could fail
Data Source
AI summary
A memory region can durably self-identify as being faulty when read. Information that would have been assigned to the faulty memory region can be assigned to another of that sized region in memory using a replacement encoding technique. For phase change memory, at least two fault states can be provided for durably self-identifying a faulty memory region; one state at a highest resistance range and the other state at a lowest resistance range. Replacement cells can be used to shift or assign data when a self-identifying memory fault is present. A memory controller and memory module, alone or in combination may manage replacement cell use and facilitate driving a newly discovered faulty cell to a fault state if the faulty cell is not already at the fault state.