Memory Scrub Error Detection via Dynamic Mode Switching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory devices struggle to detect multiple-bit errors during scrub operations, often relying on single error correction (SEC) schemes that fail to correct multiple-bit errors and may introduce additional errors.
Innovation Solution
Implementing a configurable error control mode that allows memory devices to switch between single error correction (SEC) and single error correction double error detection (SECDED) schemes, enabling detection of multiple-bit errors by reading both first and second error control information from different partitions of the memory array.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single error correction (SEC) scheme is used during scrub operations, then the device complexity is reduced and operation is simpler, but multiple-bit errors cannot be detected and may introduce additional errors
Solution Approach 1:
The memory device dynamically switches between SEC and SECDED error control modes based on the scrub operation status. During normal operations, the simpler SEC mode is used. During scrub operations, the system transitions to the more capable SECDED mode to detect multiple-bit errors, then reverts to SEC mode afterward. This dynamic adaptation resolves the contradiction by applying complexity only when needed for error detection.
Solution Approach 2:
The system changes the error control parameter (error detection capability) by switching between two predefined modes: SEC mode with basic correction capability and SECDED mode with enhanced multiple-bit error detection capability. This parameter change allows the system to optimize reliability during scrub operations while maintaining simpler operation during normal functions, resolving the contradiction between reliability and complexity.
2Reliability
If a single error correction (SEC) scheme is used, then the operation speed is faster with fewer processing steps, but data integrity deteriorates as multiple-bit errors go undetected
Solution Approach 1:
The system dynamically adjusts the error control mechanism based on operational context. During scrub operations where data integrity is critical, the system activates SECDED mode with comprehensive error detection. During normal operations, it uses lighter SEC mode for faster processing. This dynamic switching resolves the contradiction between reliability and productivity by applying enhanced error checking only when necessary.
3Productivity
If single error correction is performed during scrub operations, then the processing overhead is minimized, but undetected multiple-bit errors may lead to additional errors being introduced
Solution Approach 1:
The system performs a preliminary switch to SECDED mode before executing the scrub operation, ensuring that multiple-bit errors can be detected during the critical data verification process. After the scrub operation completes, the system reverts to SEC mode. This preliminary action of mode switching ensures error detection reliability is established before the scrub operation begins, resolving the contradiction between productivity and reliability.
Data Source
AI summary
Methods, systems, and devices for multiple bit error detection in scrub operations are described. A memory device may initiate a scrub operation on a set of rows of the memory device and determine whether to perform the scrub operation using a first error control mode associated with correcting single-bit errors or using a second error control mode associated with correcting single-bit errors and detecting multiple-bit errors. In a case that the memory device determines to perform the scrub operation using the second error control mode, the memory device may read data and first error control information from the set of rows and additional second error control information for a different partition of the memory device storing error control information. The memory device may then correct single-bit errors and detect multiple-bit errors based on the data, first error control information, and second error control information as part of the scrub operation.