NAND Flash Memory Buffer Utilization via Data Duplication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In nonvolatile memory systems, inflexible host buffers that are smaller than the capacity of the characteristic portion of the NVM lead to underutilization and reliability issues, as they fail to fully utilize the storage capacity, resulting in wasted space and potential data integrity concerns.
Innovation Solution
The system employs a method where data is written in packets matching the characteristic portion capacity, and if the host buffer is smaller, additional data such as duplicates or whitened data is programmed to fill the unused capacity, ensuring maximum utilization and reliability by duplicating write requests or filling the unutilized space with metadata or whitened data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the host buffer capacity is kept small to simplify the host device architecture, then the device complexity is reduced, but the memory utilization and reliability of the NVM deteriorate
Solution Approach 1:
The patent applies copying by duplicating the data written to the NVM and storing copies in the host buffer. This allows the host buffer to be smaller while still ensuring data integrity and reliability, as multiple copies of the data are maintained across different storage locations.
Solution Approach 2:
The patent implements preliminary action by pre-writing data to the NVM before the host buffer is fully processed. The system writes data to the NVM and then fills the host buffer with additional data or metadata, ensuring that the NVM is fully utilized and data integrity is maintained even with a smaller host buffer.
2Productivity
If the host buffer capacity is increased to match the NVM characteristic portion capacity, then the memory utilization is maximized, but the host device complexity and cost increase
Solution Approach 1:
Instead of increasing the host buffer capacity to match the NVM characteristic portion capacity, the patent uses copying by duplicating data and storing it in the host buffer. This achieves high memory utilization in the NVM while keeping the host buffer smaller and less complex.
Solution Approach 2:
The patent applies partial action by writing only the necessary data to the NVM and using the host buffer for additional purposes such as storing metadata or duplicate data. This allows the system to achieve high memory utilization without requiring the host buffer to be fully sized to match the NVM capacity.
3Reliability
If additional data is programmed to the characteristic portion to avoid underutilization, then the memory utilization and reliability improve, but the write operation complexity increases
Solution Approach 1:
The patent simplifies the write operation complexity by using copying - the host buffer stores duplicate copies of the data that were written to the NVM. This allows additional data to be programmed to the characteristic portion without significantly increasing the complexity of the write operation, as the duplication is handled efficiently.
Solution Approach 2:
The system implements self-service by automatically managing the additional data programming to the characteristic portion. The host controller and memory controller work together to fill the host buffer with appropriate data after the initial write to the NVM, reducing the manual intervention and complexity required.
Data Source
AI summary
Systems and methods for writing to high-capacity memory are disclosed. In high-capacity memory systems in which the capacity of the characteristic portion of the memory (e.g., a page of NAND flash memory) exceeds the capacity of a buffer used to write to the memory, underutilization issues are prevalent. Data organized in the buffer can be combined with additional data to improve utilization of the characteristic portion. According to various embodiments, the additional data can include duplicate copies of the data, whitened data, or any other suitable type of data.