Memory Controller Journal Log Checkpointing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage devices face performance degradation and power wastage due to journaling, especially during unexpected events like sudden power-offs or kernel panics, as they struggle to maintain consistency between metadata and journal logs.
Innovation Solution
A method and system where a memory controller manages metadata updates based on the latest journal log after a checkpoint, allowing for efficient deletion of old journal logs and ensuring consistency by rebuilding metadata using journal logs and old metadata after unexpected events, thereby minimizing performance drops and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If journaling is implemented to maintain data consistency during unexpected events, then reliability is improved, but write performance deteriorates due to continuous metadata updates and power consumption increases
Solution Approach 1:
The patent segments the journal log management into transaction-based groups, where journal logs are processed collectively rather than individually. This segmentation allows the system to maintain consistency checks only at transaction boundaries, reducing the frequency of metadata updates while ensuring data integrity across all journal entries within each transaction group.
Solution Approach 2:
The patent extracts the consistency verification process from every write operation and concentrates it only at checkpoint events. By taking out the continuous consistency checking from the normal write path and performing it only at designated checkpoint moments, the system maintains reliability while eliminating performance overhead during regular write operations.
2Reliability
If journal logs are continuously updated to reflect metadata changes, then data consistency is improved, but power consumption increases due to unnecessary updates
Solution Approach 1:
The patent implements periodic checkpoint events where metadata consistency is verified and journal logs are updated only at these periodic intervals rather than continuously. This periodic action ensures that consistency is maintained while eliminating the continuous power consumption associated with frequent metadata update operations.
Solution Approach 2:
The patent applies local quality by performing consistency verification and journal log updates only at specific checkpoint locations in the write operation flow, rather than uniformly across all operations. This localized approach concentrates computational effort where it is most needed for reliability while minimizing energy consumption in other parts of the system.
3Reliability
If metadata is updated for every journal log entry, then data consistency is improved, but device complexity increases due to frequent synchronization operations
Solution Approach 1:
The patent merges multiple individual journal log updates into a single collective update operation performed at checkpoint events. Instead of synchronizing metadata with each individual journal log entry, the system combines all journal logs within a transaction group and performs a unified metadata update, thereby reducing synchronization complexity while maintaining consistency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed are a storage device (1000) including a memory controller (100) and a method of operating the memory controller. A storage device according to the technical idea of the present disclosure includes a write buffer for storing write data that is not grouped into a transaction, a non-volatile memory device (200) including a journal buffer where journal logs are stored, a volatile memory device (300) for temporarily storing first metadata (320), and a memory controller for updating the first metadata to the second metadata based on the journal log stored after the start of the checkpoint among the journal logs stored in the journal buffer.