Storage Controller Data Migration via Compressed Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing techniques for predicting the average number of deletions in SSDs for data migration in storage systems using gross write quantity are insufficient due to divergence between uncompressed data stored by the storage controller and compressed data in SSDs, leading to inaccurate data migration and increased load on the storage controller.
Innovation Solution
A storage system with multiple nonvolatile semiconductor memory chips and a memory controller that compresses data and transmits relationship information between write data and compressed data quantities, allowing the storage controller to estimate and balance data distribution across SSDs, thereby improving the accuracy of data migration and reducing the load on the storage controller.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the storage controller uses gross write quantity to predict average number of deletions for data migration, then the data migration process can be simplified, but the accuracy of data distribution balancing deteriorates due to divergence between uncompressed and compressed data quantities
Solution Approach 1:
The storage controller obtains feedback information from each SSD about the actual write quantity to compressed data, and uses this feedback to correct the prediction of average number of deletions. This feedback mechanism resolves the contradiction by providing accurate data for migration decisions without requiring complex preprocessing of compression ratios.
Solution Approach 2:
The system changes the parameter used for prediction from gross write quantity alone to a corrected value that accounts for compression ratios. By adjusting this parameter based on actual compressed data quantities, the system achieves accurate data distribution balancing while maintaining a simple migration process.
2Measurement precision
If the storage controller obtains detailed information from each SSD to improve migration accuracy, then data distribution balancing improves, but the load on the storage controller increases
Solution Approach 1:
The patent extracts only the necessary feedback information (actual write quantity to compressed data) from each SSD, rather than obtaining all possible detailed information. This selective extraction maintains migration accuracy while minimizing the processing load on the storage controller.
3Quantity of substance
If data compression is applied in SSDs to increase storage capacity, then the storage capacity increases, but the accuracy of deletion prediction based on gross write quantity deteriorates
Solution Approach 1:
The system uses feedback from SSDs about actual compressed data write quantities to correct deletion predictions. This feedback compensates for the distortion introduced by compression, allowing accurate prediction despite the increased storage capacity enabled by compression.
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 enhances the accuracy of data distribution balancing across SSDs, extending the overall storage system's lifetime by accurately identifying and migrating data to optimize SSD usage.
Implementation Method 1
converts write data to compressed data by compressing the write data
Data Source
AI summary
The storage system includes a plurality of storage devices and a storage controller. The storage controller stores a data request quantity indicating the data quantity of write data written to the target area in a specific period, and estimates, based on the quantity of request data and relationship information received from storage devices, the estimated data quantity written to the nonvolatile semiconductor memory chips based on the write data written to the target area in the specific period. The storage controller selects a second logical storage area with an estimated data quantity less than an estimated data quantity for the first logical storage area and assigned to a storage device different from a storage device assigned to the first logical storage area, and migrates the first data stored in the first logical storage area to the second logical storage area.