Storage Device HMB Integrity Checking and Address Remapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage devices face challenges in preventing and recovering data corruption caused by hardware failures in host memory buffers, leading to repetitive recovery operations and increased recovery overhead.
Innovation Solution
Incorporating an integrity checking module and an HMB mapping module that checks data integrity and maps corrupted host memory buffer addresses to alternative addresses within the host memory or internal memory, reducing recovery overhead by avoiding locations with hardware errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the storage device executes recovery operations on corrupted data in the HMB, then data integrity is restored, but recovery overhead increases and repetitive corruption occurs when the host memory has hardware failures
Solution Approach 1:
The storage device proactively detects corrupted data in the HMB and performs recovery operations before the corruption affects system operation. By executing recovery in advance and remapping corrupted addresses, the system prevents repetitive corruption cycles and reduces the time loss associated with repeated recovery attempts.
Solution Approach 2:
The storage device introduces an address remapping mechanism that acts as an intermediary between the host memory buffer and the storage operations. When corruption is detected at a particular HMB address, the system remaps that address to an alternative location, preventing direct access to corrupted data and eliminating the need for repetitive recovery operations.
2Quantity of substance
If the storage device uses host memory buffer for data storage, then storage capacity is extended, but data corruption risk increases due to host memory hardware failures
Solution Approach 1:
The storage device implements an address remapping mechanism that serves as an intermediary layer between the host memory buffer and storage operations. When data corruption is detected at a specific HMB address, the system remaps that address to an alternative location, preventing direct access to corrupted data while maintaining the extended storage capacity provided by the HMB.
Solution Approach 2:
The storage device creates redundant copies of data in alternative memory locations within the HMB. When corruption is detected at the original address, the system can redirect access to the copied data at the alternative address, thereby maintaining data availability and preventing corruption propagation while preserving the extended storage capacity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a storage device. The storage device includes an integrity checking module checking integrity of data stored in a first host memory buffer (HMB) address of an HMB in a host coupled to the storage device, and an HMB mapping module mapping, if the integrity checking module determines the data as corrupted, the first HMB address to a second address.