Read Disturb Temperature Identification in Storage Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional read temperature identification systems in information handling systems are costly, complex, and inaccurate, requiring dedicated host memory and power-fail safety measures, and often introduce performance issues due to host processor involvement, with limited granularity and resilience to changing workloads.
Innovation Solution
A read-disturb-based read temperature identification system that utilizes storage devices to generate local read temperature information and normalize it to provide global read temperature information, leveraging the read disturb effect to identify frequently accessed data without relying on host processors, thus reducing complexity and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional read temperature identification systems use host memory and host processor to track read access, then read temperature information can be identified, but system complexity and costs increase due to dedicated memory requirements and power-fail safety measures
Solution Approach 1:
The storage device autonomously generates and maintains read temperature information using its own memory and processor resources, without requiring external host system support. The storage device independently tracks read access patterns and maintains this information through its own power management mechanisms
Solution Approach 2:
The read temperature identification functionality is extracted from the host system and transferred to the storage device itself. This separates the temperature tracking function from the host memory and processor, eliminating the need for dedicated host memory and power-fail safety measures in the host system
2Loss of information
If host processor is used to generate and synchronize read temperature maps across multiple storage devices, then global read temperature information can be obtained, but system complexity increases due to synchronization requirements
Solution Approach 1:
The global read temperature information is segmented into individual storage device-specific temperature maps. Each storage device generates and maintains its own local read temperature information independently, which can then be queried and aggregated by the host system without requiring complex real-time synchronization mechanisms
Solution Approach 2:
Each storage device preliminarily generates and maintains its own read temperature map in advance, so that when the host system needs global read temperature information, the data is already prepared and available locally at each device, eliminating the need for complex synchronization during operation
3Productivity
If conventional systems allocate storage devices based on read temperature, then storage resource allocation can be optimized, but performance issues arise due to host processor involvement in tracking
Solution Approach 1:
The storage device independently tracks read access patterns and generates temperature information without burdening the host processor. This self-service approach eliminates the performance overhead associated with host processor involvement in tracking, while still enabling optimized storage resource allocation based on accurate temperature data
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables accurate and efficient identification of read temperatures across storage devices, optimizing storage resource allocation and improving performance by allowing data to be stored on appropriate devices based on access frequency, while reducing the need for costly host memory and power-fail safety measures.
Implementation Method 1
A read-disturb-based read temperature identification system that utilizes storage devices to generate local read temperature information and normalize it to provide global read temperature information, leveraging the read disturb effect to identify frequently accessed data
Data Source
AI summary
A read-disturb-based read temperature information utilization system includes a read-disturb-based read temperature information utilization subsystem coupled to a storage subsystem including storage devices that each generate local read-disturb-based read temperature information associated with that storage device. The read-disturb-based read temperature information utilization subsystem retrieves at least some of the local read-disturb-based read temperature information generated by each storage device and a number of reads associated with that storage device and, based on the number of reads associated with each of the storage devices, normalizes the at least some of the local read-disturb-based read temperature information retrieved from each of the storage devices to generate normalized local read-disturb-based read temperature information for each of the storage devices. The read-disturb-based read temperature information utilization subsystem then uses that normalized local read-disturb-based read temperature information for the storage devices to generate normalized global read-disturb-based read temperature information for the plurality of storage devices.