Storage Controller Parity Pre-calculation for Tape Read Latency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing storage controllers face inefficiencies in data read times due to delayed restoration of secondary data, which prolongs the concurrent reading and restoration process of data from magnetic tapes.
Innovation Solution
A storage controller that adds dummy data to smaller data sets to match the size of the largest data set, calculates parities based on this data, and stores these parities in a faster storage device, allowing for concurrent reading and restoration of multiple data sets without waiting for initial data retrieval from the tape.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If the storage controller waits to start restoration of second data until first data is read from magnetic tape, then data restoration accuracy is ensured, but data read time increases
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing parity data for all data sets before the read operation begins. This allows the restoration process to start concurrently with the read operation rather than waiting for the read to complete, thereby reducing total data read time while ensuring restoration accuracy through pre-prepared parity information
Solution Approach 2:
The patent implements continuity of useful action by enabling the restoration process to run concurrently with the read operation. The controller reads first data from magnetic tape while simultaneously restoring second data using pre-stored parity, maximizing resource utilization and reducing overall processing time without compromising data integrity
2Productivity
If the storage controller reads multiple data sets sequentially from magnetic tape, then data read accuracy is maintained, but processing efficiency decreases
Solution Approach 1:
The patent applies preliminary action by pre-calculating parity data for all data sets before the read operation. This preparation enables the controller to perform restoration operations concurrently with reading, transforming a sequential process into a parallel one and significantly improving processing efficiency
Solution Approach 2:
The patent segments the data into multiple data sets (first data, second data, third data) with different priorities. By dividing the data and applying different restoration strategies to different segments, the system can concurrently process multiple segments, improving overall productivity while managing the complex coordination required for concurrent operations
Data Source
AI summary
A storage controller, when writing n sets of data into a first storage device, adds dummy data to other sets of data except for a first set of data having a largest size among the n sets of data such that sizes of other sets of data become equal to the size of the first set of data, calculates (n−1) parities based on the first set of data and the other sets of data, and when reading the n sets of data from the first storage device, concurrently performs a processing of reading a second set of data having a smallest size among the n sets of data from the first storage device and a processing of restoring each of two or more sets of data in the n sets of data except for the second set of data, by using the (n−1) parities and the dummy data.