Memory Read Counter Scaling for Read Disturb Scan Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional memory sub-systems face performance degradation and increased power consumption due to inefficient tracking of read operations, leading to unnecessary data integrity scans and resource utilization, as they cannot distinguish between different types of read disturb stress, such as single wordline and uniform read disturb, resulting in excessive memory management operations.
Innovation Solution
Implementing a first-in-first-out (FIFO) read command matrix to track sequential read commands and calculate a scaled read counter increment based on time differences between commands, using a look-up table to adjust the increment amount, thereby accurately estimating latent read disturb stress and reducing the frequency of data integrity scans.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional read counter incrementing is used, then read operations are tracked, but performance degradation and increased power consumption occur due to unnecessary data integrity scans
Solution Approach 1:
The patent changes the parameter of read counter incrementing from a fixed value to a dynamic scaled value based on time differences between read commands. By calculating time differences and applying scaling factors from a look-up table, the system adjusts the increment amount to reflect actual read disturb stress, reducing unnecessary scans while maintaining data integrity detection capability
Solution Approach 2:
The system transitions from a static read counter incrementing scheme to a dynamic one that adapts based on temporal patterns of read commands. The FIFO structure stores time-stamped read commands, and the scaling mechanism dynamically adjusts increment values based on observed time differences, allowing the system to respond to varying read patterns and reduce unnecessary data integrity scans
2Reliability
If conventional read counter incrementing is used, then read operations are tracked, but excessive memory management operations occur due to inability to distinguish read disturb stress types
Solution Approach 1:
The patent introduces time difference as a new parameter for characterizing read operations. By calculating the time difference between sequential read commands and mapping it to scaling factors, the system transforms uniform read counter incrementing into differentiated incrementing that reflects actual read disturb stress levels, enabling more precise error detection without excessive management operations
Solution Approach 2:
The patent segments the read operation tracking into distinct time-based categories using the FIFO structure and time difference calculations. By dividing read operations into groups based on their temporal spacing and applying different scaling factors to each segment, the system can distinguish between different types of read disturb stress (single wordline vs. uniform) and trigger scans only when necessary
3Device complexity
If conventional read counter incrementing is used, then simple tracking is implemented, but power consumption increases due to unnecessary scans
Solution Approach 1:
The system implements a dynamic tracking mechanism that adjusts its activity level based on observed read patterns. By using the FIFO structure to monitor time differences and applying scaling factors, the system reduces the frequency of data integrity scans when read commands are spaced appropriately in time, thereby reducing power consumption while maintaining adequate tracking capability
Solution Approach 2:
The patent changes the tracking mechanism from uniform incrementing to scaled incrementing based on time parameters. This parameter change allows the system to maintain accurate tracking of read disturb stress while reducing the number of scans triggered, thereby lowering power consumption without sacrificing tracking effectiveness
Data Source
AI summary
A system includes a memory device including multiple memory cells and a processing device operatively coupled to the memory device. The processing device is to receive a first read command at a first time and a second read command at a second time. The processing device is further to determine that the first read command and the second read command are directed to an at least partially same set of memory cells. The processing device is further to determine a scaling factor corresponding to a difference between the first time and the second time. The processing device is further to increment a read counter based on the scaling factor. Responsive to determining that a value of the read counter satisfies a threshold criterion, the processing device is further to perform a media management operation with respect to the at least partially same set of memory cells.