Storage Controller Parallel Journal Replay for Metadata Restoration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semiconductor memory devices, particularly nonvolatile memory devices like flash memory, face reliability issues due to the loss of metadata such as mapping tables when power is interrupted, leading to unreliable data storage.
Innovation Solution
A storage device with a nonvolatile memory and a storage controller that generates and replays journal data in parallel to restore metadata, using a journal data generator and replayer to split and manage journal data based on dependency, ensuring reliable and efficient restoration of metadata even in power failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If journal data is replayed sequentially to restore metadata, then reliability is improved, but restoration time increases
Solution Approach 1:
The journal data is divided into multiple groups based on dependency relationships. Independent groups can be replayed in parallel, while dependent groups maintain sequential processing. This segmentation allows concurrent execution of independent restoration tasks, reducing total restoration time while maintaining reliability through proper dependency handling.
Solution Approach 2:
The replay process dynamically adjusts execution based on dependency analysis. The system identifies which journal data groups can be processed simultaneously and which must be processed sequentially, optimizing the restoration process by parallelizing independent operations while preserving the required sequential order for dependent operations.
2Productivity
If journal data is split into multiple groups for parallel replay, then restoration speed is improved, but system complexity increases
Solution Approach 1:
Journal data is segmented into groups based on dependency relationships, allowing parallel processing of independent groups. The segmentation logic analyzes dependency markers in journal data to create manageable groups that can be processed concurrently, improving restoration speed without requiring complex distributed systems.
Solution Approach 2:
The system performs preliminary analysis of journal data dependencies before replay begins. By pre-processing the journal data to identify dependency relationships and group them accordingly, the system prepares the optimization structure in advance, reducing the complexity of real-time coordination during the actual replay process.
3Reliability
If all journal data is replayed to ensure complete metadata restoration, then reliability is improved, but processing time increases
Solution Approach 1:
The journal data replay process is segmented into multiple independent groups that can be processed in parallel. By dividing the complete journal data set into concurrent processing groups, the system maintains comprehensive restoration coverage while reducing the total wall-clock time required to process all journal data.
Solution Approach 2:
The replay system dynamically manages the processing of journal data groups based on their dependency relationships. Independent groups are executed concurrently to minimize total processing time, while the system ensures that all necessary journal data is eventually processed to maintain complete metadata restoration reliability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a storage device which includes a nonvolatile memory device, and a storage controller that controls the nonvolatile memory device and updates metadata based on an operation of the nonvolatile memory device, and the storage controller includes a journal data generator that generates a plurality of journal data associated with the update of the metadata, and a journal data replayer that replays the plurality of journal data in parallel to restore the metadata.