Volatile Memory Runtime Repair via Patrol Read Error Accumulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Volatile memory devices, such as DRAM, face increasing bit errors due to shrinking fabrication design rules, leading to reduced yield and operational issues after being mounted in SSDs, as existing repair methods are ineffective for faults occurring post-manufacturing.
Innovation Solution
A method for controlling the repair of volatile memory devices through a patrol read operation to generate error position information, accumulate error data, determine error attributes, and perform runtime repairs to prevent errors from becoming uncorrectable, thereby enhancing the reliability and lifetime of the devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If fabrication design rules are shrunk to increase memory density, then storage capacity is improved, but bit error rate increases and yield decreases
Solution Approach 1:
The patent performs preliminary error detection through patrol read operations before faults become uncorrectable. By continuously monitoring memory cells during idle periods and accumulating error information, the system proactively identifies and repairs soft errors and potential hard errors before they lead to device failure, thus maintaining reliability despite continued scaling.
Solution Approach 2:
The patent implements a feedback mechanism where error information from patrol reads is accumulated and analyzed to determine error attributes. This feedback loop enables the system to adaptively identify failing memory cells and perform targeted repairs, creating a closed-loop control system that continuously improves memory reliability as cells degrade over time.
2Productivity
If conventional repair methods are used during manufacturing, then yield is improved, but faults occurring after mounting in SSD cannot be repaired
Solution Approach 1:
The patent transforms the repair capability from a static, one-time manufacturing process to a dynamic, continuous operation. The memory device performs patrol read operations and repairs throughout its operational lifetime, adapting to newly emerging faults that were not present during manufacturing. This dynamic approach enables repair of faults that occur after the device is mounted in the SSD.
Solution Approach 2:
The patent enables the memory device to perform self-diagnosis and self-repair through automated patrol read operations and error analysis. The system independently identifies failing cells, determines error attributes, and executes repairs without external intervention, allowing post-manufacturing repair capabilities that conventional methods do not provide.
3Measurement precision
If patrol read operation is performed repeatedly to detect errors, then error detection capability is improved, but device complexity increases
Solution Approach 1:
The patent makes the existing memory device perform multiple functions: normal data storage, patrol read operations for error detection, error information accumulation, and repair operations. By reusing existing memory cells and circuits for both data storage and error monitoring, the patent avoids adding significant complexity while achieving comprehensive error detection capabilities.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A method of controlling repair of a volatile memory device, includes, performing a patrol read operation repeatedly to provide error position information of errors included in read data from a volatile memory device, generating accumulated error information by accumulating the error position information based on the patrol read operation performed repeatedly, determining error attribute based on the accumulated error information, the error attribute indicating correlation between the errors and a structure of the volatile memory device, and performing a runtime repair operation with respect to the volatile memory device based on the accumulated error information and the error attribute. The errors may be managed efficiently to prevent failure of the volatile memory device, and thus performance and lifetime of the volatile memory device and the storage device may be enhanced.