Zone Memory Parity Protection for NAND Block Failures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional memory sub-systems face challenges in efficiently managing block failures during programming of multi-level cell (MLC), triple-level cell (TLC), or quad-level cell (QLC) blocks, leading to high overprovisioning of SLC cache blocks and reduced competitiveness.
Innovation Solution
The implementation of block failure protection for a memory sub-system that supports zones involves matching non-parity zones filling up at similar rates, generating parity for a stripe of data across these zones, and storing the parity in a parity zone, using a RAIN technique for error-correction and a zone binning algorithm to identify matching zones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full SLC-block caching is used for block failure protection, then reliability is improved, but device complexity and resource usage increase due to excessive SLC cache blocks required
Solution Approach 1:
The patent segments the memory sub-system into distinct non-parity zones and parity zones. Non-parity zones store user data while parity zones store protection data. This segmentation allows the system to provide block failure protection without requiring full SLC caching across all blocks, thereby reducing overprovisioning while maintaining reliability.
Solution Approach 2:
The patent applies different quality characteristics to different zones: non-parity zones use the memory type appropriate for their function (e.g., QLC, TLC, MLC), while parity zones use more reliable SLC blocks specifically for protection. This local differentiation of quality allows the system to achieve protection where needed without over-provisioning the entire system.
2Reliability
If more SLC cache blocks are allocated for protection, then reliability is improved, but resource usage increases making the system less competitive
Solution Approach 1:
The patent implements partial protection by applying parity protection only to specific zones rather than all blocks. The system uses a predetermined number of SLC blocks for parity purposes, which is fewer than the blocks protected, achieving protection without the excessive resource allocation of full SLC caching.
Solution Approach 2:
The patent changes the parameter of SLC block allocation from a fixed high ratio (full SLC caching requiring 4:1 or more SLC to QLC blocks) to a configurable ratio determined by the predetermined number of SLC blocks. This allows optimization of the balance between protection and resource usage, achieving competitive systems with reduced SLC block requirements.
3Reliability
If conventional full SLC caching is used, then block failure protection is achieved, but manufacturing cost and resource intensity increase
Solution Approach 1:
By segmenting the memory sub-system into non-parity and parity zones with different protection levels, the patent reduces the overall resource intensity compared to uniform full SLC caching. This segmented approach allows cost-effective implementation by applying protection only where necessary.
Solution Approach 2:
The patent creates parity copies of data in dedicated parity zones using SLC blocks, rather than maintaining full SLC caches for all data. This copying approach provides protection while reducing resource intensity, as only critical protection data is replicated in the more expensive SLC medium.
Data Source
AI summary
Various embodiments provide block failure protection for a memory sub-system that supports zones, such a memory sub-system that uses a RAIN (redundant array of independent NAND-type flash memory devices) technique for data error-correction. For some embodiments, non-parity zones of a memory sub-system that are filling up at a similar rate are matched together, a parity is generated for stored data from across the matching zones, and the generated parity is stored in a parity zone of the memory device.