Optimizing TRT Parameters via Selective Wordline BER Scanning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage devices face performance degradation and read timeouts due to non-optimal Thermal Region Tag (TRT) parameters, which are not periodically refreshed and fine-tuned to account for temperature variance and data retention issues.
Innovation Solution
The storage device updates and optimizes TRT parameters by selecting indicative and representative wordlines, performing bit error rate (BER) estimations, and conducting BER Estimation Scans (BES) to determine valid wordlines and obtain optimal TRT parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If TRT parameters are periodically refreshed and fine-tuned, then read accuracy and performance are improved, but device complexity and overhead increase
Solution Approach 1:
The patent performs preliminary BER estimation on indicative wordlines before conducting full BES checks on representative wordlines. This preliminary filtering action identifies valid wordlines in advance, preventing unnecessary complex operations and reducing overall system overhead while maintaining read accuracy.
Solution Approach 2:
The patent segments the wordline validation process into two distinct stages: (1) rapid BER estimation on indicative wordlines to filter out invalid candidates, and (2) comprehensive BES checks only on representative wordlines from valid indicative wordlines. This segmentation reduces the complexity of any single operation while achieving the same reliability goal.
2Measurement precision
If BER estimation and BES checks are performed on all wordlines, then optimal TRT parameters are obtained, but processing time and productivity are reduced
Solution Approach 1:
The patent applies partial action by performing BER estimation on only a selected set of indicative wordlines rather than all wordlines in the block. This partial sampling approach maintains sufficient measurement precision for TRT parameter accuracy while dramatically improving processing speed and productivity.
Solution Approach 2:
The patent performs preliminary BER estimation on indicative wordlines to identify valid candidates before conducting the more time-consuming BES checks on representative wordlines. This preliminary filtering action reduces the total number of expensive operations needed, thereby improving overall productivity while maintaining measurement precision.
3Productivity
If indicative and representative wordlines are selected without validation, then TRT update overhead is reduced, but read failures and decode errors increase
Solution Approach 1:
The patent performs preliminary BER estimation on indicative wordlines to validate their suitability before selecting them as representative wordlines for BES checks. This preliminary validation ensures that only reliable wordlines are used for TRT parameter extraction, maintaining high read success rates while keeping the process efficient.
Solution Approach 2:
The patent uses BER estimation results as feedback to determine whether indicative wordlines are valid candidates for representing blocks in BES checks. This feedback mechanism ensures that only wordlines with acceptable error rates are selected, thereby maintaining read reliability while improving update efficiency through selective validation.
Data Source
AI summary
A storage device updates optimal parameters associated with a Thermal Region Tag (TRT). A controller on the storage device assigns a TRT to blocks programmed at a given temperature range and updates an optimal TRT parameters by obtaining a set of representative wordlines and a set of indicative wordlines for a block assigned to the TRT. The controller performs a bit error rate (BER) estimation on indicative wordlines in the set until a valid indicative wordline is found. The controller determines whether a BER Estimation Scan (BES) check is to be performed when the valid indicative wordline is found. In performing the BES check, the controller performs the BER estimation on representative wordlines in the set until a valid representative wordline is found. When a valid representative wordline is found, the controller obtains the optimal TRT parameter and updates the optimal TRT parameter.