Memory Die Refresh Scheduling for Lower Peak Current Draw
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
As memory components decrease in size and density, the rate of data degradation in nearby memory cells increases due to repeated access, necessitating targeted refresh operations to prevent information loss, while simultaneous auto-refresh operations can exceed current limits and increase power consumption.
Innovation Solution
The solution involves staggering the timing of different types of refresh operations across memory dies or banks, alternating between auto-refresh, targeted refresh, and skipped refresh operations to manage current draw and reduce power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all memory dies perform auto-refresh operations simultaneously, then data integrity is maintained, but peak current draw and power consumption exceed limits
Solution Approach 1:
The patent divides the memory system into multiple independent memory dies (first memory die and second memory die), each capable of independently performing refresh operations. This segmentation allows different dies to operate at different times, preventing simultaneous peak current draws while maintaining overall data integrity across the memory system.
Solution Approach 2:
The patent implements periodic refresh operations with staggered timing between different memory dies. Instead of all dies refreshing simultaneously, the refresh operations are distributed across different time periods, reducing peak power consumption while ensuring each die maintains its data integrity through regular periodic refreshing.
2Quantity of substance
If memory density is increased by decreasing component size, then storage capacity improves, but data degradation rate increases due to repeated access
Solution Approach 1:
The patent extracts and addresses the specific problem of row hammer-induced data degradation by implementing targeted refresh operations that specifically refresh memory cells affected by repeated access patterns. This selective approach addresses the degradation issue without requiring a complete refresh of all memory cells, maintaining efficiency while improving reliability.
3Reliability
If targeted refresh operations are performed frequently, then data degradation is prevented, but power consumption increases
Solution Approach 1:
The patent applies local quality by implementing targeted refresh operations that specifically address only the memory cells requiring refresh due to row hammer effects, rather than uniformly refreshing all memory cells. This localized approach prevents data degradation in vulnerable cells while minimizing unnecessary power consumption from redundant refresh operations.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments of the disclosure are drawn to apparatuses and methods for staggering the timing of skipped refresh operations on a memory. Memory cells of memories may need to periodically perform refresh operations. In some instances, auto-refresh operations may be periodically skipped when charge retention characteristics of the memory cells of the memory exceed the auto-refresh frequency. To reduce peak current draw during refresh operations, the skipped refresh operations may be staggered across different portions of the memory. In one example, the skipped refresh operation may be staggered in time among memory dies of the memory to limit a number of memory dies that are performing an auto-refresh operation to a maximum number. In another example, the skipped refresh operation may be staggered in time among memory banks of a single memory array to limit a number of memory banks that are performing an auto-refresh operation to a maximum number.