Memory Controller Channel Error Recovery via Event Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing error handling methods for rewritable nonvolatile memory modules often result in reduced accessing performance due to the need to stop operations and check multiple failed events, leading to inefficiencies in channel recovery.
Innovation Solution
An error handling method that involves tracking failed events through counting values and recovering channel operations when the values meet a threshold criterion, allowing for immediate resumption of operations and improved performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a command is executed in a pipeline manner, then accessing efficiency is improved, but multiple failed events may occur on the channel leading to operation stoppage
Solution Approach 1:
The patent implements a feedback mechanism where the memory controlling circuit unit monitors finished events on each channel, detects failed events, and provides feedback to determine when to resume operations. The circuit tracks the number of failed events and automatically resumes channel operations when the failed event count falls below a threshold, creating a closed-loop control system that balances pipeline efficiency with error handling.
Solution Approach 2:
The patent changes the operational parameters of the channel dynamically by adjusting the operation state based on the number of failed events. When the failed event count is below the threshold, the channel operates in pipeline mode for high efficiency; when the threshold is exceeded, the channel operation is stopped for error handling. This parameter change allows the system to adapt between performance and reliability states.
2Reliability
If the operation of the channel is stopped to check failed events, then error handling is performed, but accessing performance is reduced
Solution Approach 1:
The patent applies partial action by not completely stopping channel operations whenever a failed event occurs. Instead, the channel continues to operate in pipeline mode, and only specific error handling actions are taken for the failed events. The operation is partially suspended only when the number of failed events exceeds the threshold, rather than stopping immediately at the first error, thus maintaining overall accessing performance while still handling errors.
Solution Approach 2:
The patent performs preliminary error handling by monitoring and counting failed events before completely stopping channel operations. The system prepares error handling in advance by tracking the number of failures and only interrupts the pipeline operation when necessary (when the threshold is exceeded). This preliminary monitoring approach minimizes the impact on accessing performance while ensuring errors are handled.
3Reliability
If all failed events are checked before channel recovery, then error handling is thorough, but operation recovery is delayed
Solution Approach 1:
The patent performs preliminary error handling by monitoring and counting failed events as they occur during pipeline operations. Instead of waiting for all events to complete before checking for errors, the system continuously tracks failed events and prepares for recovery in advance. When the number of failed events falls below the threshold, the channel is automatically resumed without waiting for complete error verification, reducing recovery time while maintaining reliability.
Solution Approach 2:
The patent implements dynamic error handling where the channel operation state changes dynamically based on the number of failed events. The system transitions between operating states (pipeline mode, error handling mode, recovery mode) based on real-time failure counts. This dynamic approach allows the channel to recover quickly when errors are resolved rather than waiting for static completion of all error checks, reducing recovery time while maintaining thorough error handling.
Data Source
AI summary
An error handling method, a memory storage device and a memory controlling circuit unit are provided. The method includes obtaining a finished event corresponding to a channel; determining whether the finished event is a failed event, if the finished event is the failed event; stopping an operation of the channel and performing a first update operation on a counting value corresponding to the channel; and if the finished event is not the failed event, keeping the counting value corresponding to the channel unchanged and processing the finished event. The step of the processing the finished event includes performing a second update operation on the counting value corresponding to the channel if the finished event is the failed event, and recovering the operation of the channel if the counting value matches a threshold criterion. Accordingly, it can improve the accessing performance.