Adaptive Data Backup for Non-Volatile Multi-Level Cell Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The performance of write operations in non-volatile multi-level cell (MLC) NAND flash memory devices is degraded due to inefficient backup methods, particularly the post-backup method for paired pages, which results in long backup times and increased write overhead.
Innovation Solution
An adaptive data backup method is introduced that selects among different backup types based on the backup data size and the number of times data pages are backed up, utilizing interleaving, copy-back, and parity page pre-backup techniques to optimize the backup process by parallel processing and reducing redundant data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If post-backup method is used for paired pages, then data backup is performed, but backup time is long and write overhead increases
Solution Approach 1:
The patent applies preliminary action by performing backup operations before the main write operation completes. The controller identifies paired pages that need backup and performs the backup to a backup block in advance, rather than waiting for the main write to finish and then performing backup sequentially. This overlapping execution reduces the total backup time and write overhead.
2Reliability
If conventional backup method is used, then data is backed up to paired page, but write overhead occurs due to backup of data
Solution Approach 1:
The patent merges the backup operation with the main write operation by performing both operations simultaneously using multiple channels. The controller divides the backup data into multiple parts and transmits them through different channels in parallel, combining the backup function with the write function execution. This parallel execution eliminates the sequential overhead and improves write operation performance.
3Speed
If data is backed up using multiple channels, then backup speed increases, but channel resources are consumed
Solution Approach 1:
The patent applies partial action by dynamically selecting the number of channels to use for backup based on the amount of data to be backed up and the current system state. For smaller backup operations, fewer channels are utilized, while for larger operations, more channels are engaged. This adaptive approach optimizes backup speed while avoiding unnecessary consumption of channel resources for small backup tasks.
Data Source
AI summary
In one example embodiment of the inventive concepts, an adaptive data backup method performed in a memory system including a non-volatile multi-level cell memory device includes receiving a write command from a host and determining a backup data size which is a size of data to be backed up among data requested to be written in the write command. The adaptive data backup method further includes selecting a backup type among at least two different backup types, based on the backup data size and backing up the data according to the selected backup type.