Request-Based Refresh for DRAM Weak Rows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic random access memory (DRAM) devices face limitations in preventing data loss due to varying reasons for data retention failure, as current refresh designs are generalized and may not effectively address specific data retention issues in all memory cells.
Innovation Solution
A memory device and system that implement a request-based refresh method, where the memory device determines the need for supplemental refresh on weak rows by counting activated rows, generates a request signal for a refresh command, and performs request-based refresh in addition to normal and target refresh cycles to enhance data retention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a generalized refresh design is used, then the refresh operation can be applied to all memory cells uniformly, but it cannot effectively address specific data retention issues in all memory cells due to varying reasons for data loss
Solution Approach 1:
The refresh operation is segmented into multiple types: normal refresh, target refresh, and request-based refresh. Each type addresses different data retention needs, allowing the system to selectively apply appropriate refresh strategies to different memory rows based on their specific requirements
Solution Approach 2:
Different refresh strategies are applied to different memory rows based on their local characteristics. Weak rows that exhibit data retention issues are identified and subjected to request-based refresh, while other rows continue with normal or target refresh operations
2Reliability
If request-based refresh is performed on weak rows, then data loss is reduced by proactively identifying and refreshing problematic rows, but the device complexity increases due to additional monitoring and control logic
Solution Approach 1:
The memory device performs self-diagnosis by monitoring its own operation status and automatically generating request signals for refresh operations when weak rows are detected, reducing the need for external intervention and simplifying the overall control architecture
Solution Approach 2:
The system implements feedback mechanisms where the memory device monitors its own refresh needs and communicates with the memory controller through request signals. The memory controller responds by issuing appropriate refresh commands, creating a closed-loop control system that adapts to real-time memory conditions
3Reliability
If multiple refresh types (normal refresh, target refresh, and request-based refresh) are implemented, then data retention is improved through comprehensive coverage, but the operation time increases due to repeated refresh cycles
Solution Approach 1:
Instead of uniformly applying extensive refresh operations to all memory rows, the system performs partial refresh actions only on identified weak rows that actually need them. This selective approach reduces unnecessary refresh operations and their associated time overhead while maintaining data retention reliability
Data Source
AI summary
Provided are a memory device and a memory system performing request-based refresh, and an operating method of the memory device. The operating method includes: determining a weak row by counting an activated number of at least one row; requesting for refresh on the weak row based on a result of the determining; and performing target refresh on the weak row upon receiving a refresh command according to the requesting.