Zoned Namespace Grouping for NAND Erase Block Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ZNS technology faces challenges in aligning zone sizes with different NAND erase block sizes across various types and capacities, leading to inefficiencies such as write amplification, wasted capacity, and high metadata overhead, especially when transitioning from HDDs to SSDs in hyperscale storage systems.
Innovation Solution
Implement zone grouping technology that aligns zone sizes by determining a smallest aligned boundary based on multiple zone size-dependent parameters, including preferred zone size, erase block sizes, and namespace capacities, creating consistent zone groups that facilitate seamless data transfer and efficient erasure coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If zone sizes are aligned with different NAND erase block sizes across various types and capacities, then capacity utilization is optimized, but device complexity increases due to multiple parameters and boundaries to manage
Solution Approach 1:
The namespace is divided into multiple zones that are grouped into zone groups, where each zone group contains a specific number of zones (e.g., 4 zones per group). This segmentation allows the system to manage zone sizes in discrete units that align with NAND erase block sizes, optimizing capacity utilization while maintaining manageable complexity through structured organization.
Solution Approach 2:
The system dynamically adjusts zone group size based on the number of zones in the namespace. When the number of zones is divisible by 4, zone groups of 4 zones are used; otherwise, zone groups of 2 zones are used. This parameter change approach allows flexible adaptation to different namespace configurations while maintaining alignment with NAND erase block sizes.
2Loss of energy
If sequential writes are required to zones, then write amplification is reduced, but productivity decreases due to restricted write patterns
Solution Approach 1:
The system performs preliminary actions by requiring host systems to issue reset commands before writing to a zone, and by organizing zones into groups that can be managed as units. This preliminary organization allows the SSD to optimize write operations internally while maintaining the sequential write requirement, reducing write amplification without completely blocking productivity through efficient resource preparation.
Solution Approach 2:
The zone grouping structure acts as an intermediary layer between the host system and the physical NAND structure. This intermediate organization allows the system to enforce sequential write patterns at the zone level while providing flexibility at the zone group level, managing the trade-off between write amplification reduction and productivity maintenance.
3Manufacturing precision
If zone sizes are fixed, then manufacturing precision is improved, but adaptability decreases when transitioning from HDDs to SSDs
Solution Approach 1:
The zone grouping mechanism provides a universal interface that works across different NAND types and capacities. By organizing zones into groups with configurable sizes (2 or 4 zones per group), the system maintains consistent alignment with NAND erase block sizes regardless of the specific NAND technology being used, achieving both manufacturing precision and adaptability.
Solution Approach 2:
The system dynamically adjusts zone group configuration based on the specific NAND device being used. The controller can determine the appropriate zone group size based on the number of zones and the characteristics of the NAND media, allowing fixed-size zones to adapt to different NAND types and capacities while maintaining precise alignment.
Data Source
AI summary
An embodiment of an electronic apparatus may include one or more substrates, and logic coupled to the one or more substrates, the logic to control access to two or more persistent storage devices, wherein at least one of the two or more persistent storage devices is to be logically organized with a namespace divided into two or more zones, determine two or more different zone size dependent parameters associated with the two or more persistent storage devices, determine a smallest aligned boundary based on each of the two or more different zone size dependent parameters, and set a zone group size for access to the two or more persistent storage devices based on the determined smallest aligned boundary. Other embodiments are disclosed and claimed.