SAT Bad Block Mapping for UECC Jumboblock Reads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices face performance issues due to bandwidth-intensive scanning for uncorrectable error correction code (UECC) errors, leading to timeouts and read-only modes, as controllers need to scan storage address tables (SAT) for UECC errors, impacting read performance.
Innovation Solution
Link jumboblocks (JBs) with UECC errors to a designated 'bad block' in the SAT, updating mappings to point to this block, and return a UECC message when a read command is received, optimizing address translation and eliminating the need for separate UECC list scanning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the controller scans the SAT to determine which LBA is associated with the failed JBA and marks the respective marking as having a UECC error, then the UECC error can be identified and handled, but the scanning operation requires reading control data from the memory device and scanning the control data, which may cause a timeout to occur and is bandwidth expensive
Solution Approach 1:
The controller performs preliminary action by scanning the SAT during normal operation to identify LBAs associated with failed JBAs and marking them with UECC errors before actual read operations occur. This advance preparation eliminates the need for time-consuming scanning during error handling operations, preventing timeouts while maintaining reliable error identification.
2Reliability
If the controller saves a JB identification (ID) associated with the source JB in a UECC list, then the JB can be tracked for future operations, but the JBs associated with the JB IDs stored in the UECC list may not be allocated in future operations, which may cause the data storage device to enter a read-only mode
Solution Approach 1:
The controller extracts the LBA information from the SAT entries corresponding to failed JBAs and creates a separate UECC error list containing only the LBAs with uncorrectable errors. This extraction approach allows the system to track problematic JBAs for future operations while maintaining the flexibility to allocate other JBAs normally, preventing read-only mode by separating error tracking from allocation management.
3Reliability
If the controller scans the UECC list after each address translation from the SAT to determine if the JB ID exists in the UECC list, then UECC errors can be detected, but read performance may be impacted
Solution Approach 1:
The controller performs preliminary action by pre-scanning the SAT during normal operation to identify and mark LBAs with UECC errors before actual read operations occur. This advance preparation eliminates the need for time-consuming scanning during error handling operations, preventing timeouts while maintaining reliable error identification.
4Productivity
If the controller updates a mapping of the SAT to point to a bad block for JB with UECC error, then read and write operations are optimized by reducing scanning overhead, but the controller needs to scan the SAT to determine that the JB identification points to the bad block
Solution Approach 1:
The controller performs preliminary action by scanning the SAT during normal operation to identify LBAs associated with failed JBAs and marking them with UECC errors before actual read operations occur. This advance preparation eliminates the need for time-consuming scanning during error handling operations, preventing timeouts while maintaining reliable error identification.
Data Source
AI summary
Jumboblocks (JBs) having an uncorrectable error correction code (UECC) error are linked to a bad block in a storage address table (SAT) in order to optimize read and write operations. Bad blocks are designated as a bad block during a manufacturing process or associated with a write uncorrectable error. When a JB is identified as having the UECC error, the controller updates a corresponding mapping of the SAT to point to a bad block, where each JB having the UECC error points to the bad block. When a read command is received by the controller for the data of the JB having the UECC error, the controller scans the SAT, determines, from the SAT, that the JB identification points to the bad block, and returns a UECC message to the requester of the data of the JB having the UECC error.