Flexible Address Swap Column Redundancy for Memory Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory sub-systems face inefficiencies in addressing defects and errors due to inadequate redundancy schemes, where some regions exhaust available redundant bytes while others have unused capacity, leading to inefficient resource utilization and potential data loss.
Innovation Solution
A flexible address swap column redundancy scheme is implemented, allowing the memory array to remap memory addresses from regions with exhausted redundant locations to regions with available redundant locations, thereby optimizing the use of redundant memory locations across the memory array.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fixed redundancy schemes are used, then each address range has dedicated redundant bytes, but some regions exhaust available redundant bytes while others have unused capacity
Solution Approach 1:
The patent applies universality by making redundant memory locations serve multiple address ranges. Instead of dedicating specific redundant bytes to specific address ranges, the system allows any available redundant location to be used by any address range that needs defect repair, maximizing the utility of limited redundant resources
Solution Approach 2:
The patent implements dynamics through the dynamic remapping mechanism. When a defect is detected in an address range, the system dynamically identifies and remaps the defective address to an available redundant location, adapting the redundancy allocation in real-time based on actual defect patterns rather than using static pre-assigned redundancy
2Reliability
If more redundant memory locations are allocated to each address range, then defect repair capability improves, but memory array density and capacity are reduced
Solution Approach 1:
By making redundant locations universal resources that can serve any address range, the system achieves high reliability without increasing the total number of redundant locations. Each redundant byte can be shared across multiple address ranges, eliminating the need to allocate dedicated redundant space to each range and preserving memory array density
3Productivity
If address remapping is implemented, then efficient use of redundant locations is achieved, but additional circuitry and complexity are required
Solution Approach 1:
The patent applies self-service by implementing defect detection and remapping functionality directly within the memory array structure. The memory system performs its own defect detection during operations and automatically remaps defective addresses using available redundant locations, eliminating the need for external complexity while achieving high redundancy utilization efficiency
Data Source
AI summary
A memory device includes a memory array includes memory cells grouped into one or more address ranges. Control logic is coupled to the memory array and configured to detect one or more errors associated with one or more stored data items corresponding to a first address range of one or more address ranges. The control logic can determine that a number of the one or more stored data items exceeds a number of redundant memory locations for the first address space. Control logic can remap an association of a first memory address of at least one of the stored data items from a first address within the first address space to a second address in a second address range, where the second address range includes one or more available redundant memory locations.