Dynamic Scrubbing Controller for Heterogeneous Memory Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data centers face inefficiencies in managing and processing large-scale data due to the lack of modularized and adaptable storage resources, which are essential for emerging technologies like AI, autonomous driving, and virtual reality, as they require high data rate transmission and analysis.
Innovation Solution
A memory system comprising a group of memory devices with different types, controlled by a scrubbing controller that collects health information to determine individual scrubbing intervals based on deterioration degree, reducing intervals proportionally as deterioration increases, thereby maintaining data reliability and reducing power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed scrubbing intervals are used for all memory devices, then simplicity of control is maintained, but power consumption increases and data reliability decreases for deteriorated devices
Solution Approach 1:
The scrubbing interval is changed from a fixed value to a dynamic value that adjusts based on the health status of each memory device. The controller determines different scrubbing intervals for different memory devices according to their deterioration degrees, allowing the system to optimize between power consumption and data reliability dynamically.
Solution Approach 2:
The scrubbing interval parameter is modified based on health information parameters. The controller changes the scrubbing interval parameter according to the deterioration degree of each memory device, enabling adaptive adjustment of the scrubbing frequency to match the actual condition of the memory devices.
2Reliability
If scrubbing intervals are reduced for deteriorated memory devices, then data reliability improves, but control complexity increases
Solution Approach 1:
The controller obtains health information from each memory device and uses this feedback to determine appropriate scrubbing intervals. The health information serves as feedback about the deterioration degree, allowing the controller to adjust scrubbing intervals automatically without requiring complex manual control.
Solution Approach 2:
Each memory device effectively manages its own scrubbing needs by providing health information that reflects its deterioration degree. The controller uses this self-reported information to determine scrubbing intervals, allowing the memory devices to self-regulate their maintenance requirements based on their actual condition.
3Productivity
If health information collection and dynamic interval determination are implemented, then power consumption is optimized and reliability improved, but system complexity increases
Solution Approach 1:
The memory system is segmented into multiple independent memory devices, each with its own health information and scrubbing interval. The controller manages each memory device separately based on individual health status, allowing granular optimization without requiring complex system-wide changes.
Solution Approach 2:
The controller proactively obtains health information from memory devices before deterioration becomes critical. By collecting health information in advance and determining scrubbing intervals based on this preliminary data, the system can prevent errors before they occur, improving productivity while maintaining manageable complexity.
Data Source
AI summary
A memory system includes a memory group including a plurality of memory devices having two or more different types; and a controller configured to control data input and output for the memory group, wherein the controller includes a scrubbing controller configured to collect health information to which a deterioration degree for each of the plurality of memory devices is reflected and determine a scrubbing interval for each of the plurality of memory devices based on the health information, the scrubbing interval being reduced in proportion to the deterioration degree.