Endurance-Aware Data Placement in Mixed Media Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional storage management technologies are not cognizant of media characteristics, leading to performance and endurance issues in NAND flash storage systems, particularly for write-intensive workloads, which can result in device burn-out and data integrity problems due to improper data placement across different media types.
Innovation Solution
A data placement system that utilizes metrics such as write reduction and write invalidation factors to dynamically allocate data across 3D crosspoint and NAND flash media, employing a closed-loop feedback mechanism to optimize performance and endurance by classifying workloads and determining the best media types for data placement based on fitness criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data is placed on higher capacity media (QLC, TLC), then storage capacity increases, but endurance and performance decrease
Solution Approach 1:
The storage system is segmented into multiple media types (SLC, MLC, TLC, QLC) with different characteristics. The system divides data placement decisions based on workload characteristics, placing different portions of data on different media types to simultaneously achieve high capacity and high endurance.
Solution Approach 2:
Different portions of the storage system are assigned different qualities based on local needs. High-endurance media (SLC, MLC) are used for frequently written data, while high-capacity media (TLC, QLC) are used for less frequently written data. This local differentiation allows the system to optimize both endurance and capacity simultaneously.
2Device complexity
If conventional storage management is used without media awareness, then system complexity is reduced, but performance and endurance deteriorate due to improper data placement
Solution Approach 1:
The system implements feedback mechanisms that monitor workload characteristics, media health status, and performance metrics. Based on this feedback, the storage manager dynamically adjusts data placement decisions to optimize both performance and endurance while maintaining media awareness.
Solution Approach 2:
The storage management system transitions from static to dynamic operation. Data placement is not fixed but adapts based on changing workload characteristics, media conditions, and performance requirements. This dynamic approach allows the system to respond to changing conditions without requiring complex manual intervention.
3Quantity of substance
If more bits per cell are stored (QLC vs SLC), then capacity increases, but write performance and endurance decrease
Solution Approach 1:
The system segments write operations based on their characteristics and destinations. Write-intensive data is directed to SLC or MLC media where write performance is superior, while write-light data is placed on QLC media to maximize capacity utilization. This segmentation allows the system to achieve high overall capacity while maintaining high write performance for critical data.
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 manage access to a storage system that includes a first persistent storage device and a second persistent storage device, capture input/output telemetry for a workload on the storage system, determine one or more write reduction factors and one or more write invalidation factors for the workload based on the captured input/output telemetry, and allocate storage for the workload between the first persistent storage device and the second persistent storage device based on the one or more write reduction factors and the one or more write invalidation factors. Other embodiments are disclosed and claimed.