Compressed Repair Codes for Scalable Non-Volatile Memory Repair
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing implementations of non-volatile memory systems for memory repair in integrated circuits do not scale well with increasing device densities and decreasing sizes, leading to inefficient use of memory space due to fixed-sized repair codes and unused bits.
Innovation Solution
A method and system for compressing and decompressing repair codes using a non-volatile memory controller and logic that assigns compression parameters based on memory instance configurations, storing only usable bits and adding compression control data to reduce storage requirements, allowing for lossless compression and incremental memory repair.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed-sized repair codes are used in non-volatile memory, then memory repair functionality is provided, but storage space is inefficiently used due to unused bits
Solution Approach 1:
The patent applies parameter changes by transitioning from fixed-sized repair codes to variable-length compressed repair codes. The repair code size is dynamically adjusted based on the actual number of defective elements detected in each memory instance, eliminating the need to store unused bits while maintaining complete repair functionality.
Solution Approach 2:
The patent extracts and stores only the usable repair signature bits that correspond to actual defective elements, removing unnecessary padding bits from the fixed-sized repair codes. This extraction process reduces the storage requirement from a fixed maximum size to a variable size that matches the actual repair needs of each memory instance.
2Reliability
If non-volatile memory size is increased to accommodate all repair codes, then all memory instances can be repaired, but device area increases
Solution Approach 1:
The patent changes the parameter of repair code size from a static maximum value to a dynamic variable length based on actual defect patterns. This allows the non-volatile memory to be sized according to the average compressed repair code size rather than the maximum possible size, reducing the required device area while maintaining the ability to repair all memory instances.
Solution Approach 2:
The patent introduces dynamics into the repair code storage system by using variable-length codes that adapt to the specific defect patterns of each memory instance. The compression ratio and code length dynamically adjust based on the number and distribution of defective elements, optimizing the balance between repair coverage and device area.
3Quantity of substance
If compression is applied to repair codes, then storage requirements are reduced, but additional logic for compression and decompression is required
Solution Approach 1:
The patent applies preliminary action by pre-computing the compression parameters and control data during the memory testing and repair code generation phase. The compression algorithm and control information are prepared in advance, allowing the actual compression and decompression operations to be efficiently executed without adding significant runtime complexity.
Data Source
AI summary
One example includes an integrated circuit (IC). The IC includes non-volatile memory and logic. The logic is configured to receive repair code associated with a memory instance and assign a compression parameter to the repair code based on a configuration of the memory instance. The logic is also configured to compress the repair code based on the compression parameter to produce compressed repair code and to provide compressed repair data that includes the compressed repair code and compression control data that identifies the compression parameter. A non-volatile memory controller is coupled between the non-volatile memory and the logic. The non-volatile memory controller is configured to transfer the compressed repair data to and/or from the non-volatile memory.