Two-Dimensional Data Transform Device Storage Access Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current two-dimensional transform methods, such as two-dimensional Fourier and Hadamard transforms, face inefficiencies due to high numbers of accesses to auxiliary storage devices, which significantly impact processing time.
Innovation Solution
A two-dimensional data transform device and method that optimizes data access by reading and writing column groups from an auxiliary storage device, performing one-dimensional transforms, and switching data formats to reduce the number of accesses and enhance processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional two-dimensional transform methods are used with multiple row and column transforms, then transform accuracy is maintained, but the number of auxiliary storage device accesses increases significantly
Solution Approach 1:
The patent segments the two-dimensional transform into distinct row transform and column transform phases, processing data in organized blocks. By dividing the M×N data matrix into manageable segments and processing them systematically, the method maintains transform accuracy while optimizing access patterns to reduce total access operations.
Solution Approach 2:
The patent performs preliminary row transforms on all rows before conducting column transforms. This preliminary action organizes data in the auxiliary storage device in a way that facilitates efficient column access later, reducing the total number of access operations needed while ensuring accurate transform results.
2Reliability
If data is read and written frequently to auxiliary storage device during transform operations, then processing completeness is ensured, but processing speed decreases
Solution Approach 1:
The patent implements continuous processing by performing row transforms on all rows first, then immediately proceeding to column transforms without unnecessary intermediate reads or writes. This continuity ensures that once data is loaded into main storage, it is processed completely through both transform stages, maintaining reliability while minimizing auxiliary storage access frequency.
Solution Approach 2:
The patent uses main storage device as an intermediary buffer between auxiliary storage and processing units. Data is read from auxiliary storage to main storage once, processed through row and column transforms using the main storage as workspace, then results are written back. This intermediary approach ensures complete processing while reducing direct auxiliary storage access frequency.
3Stability of the object's composition
If traditional row-major data access patterns are used, then data continuity is maintained, but the number of auxiliary storage accesses for column operations increases
Solution Approach 1:
The patent changes the access dimension by performing all row transforms first (accessing data row-by-row in the natural order), then performing column transforms (accessing data column-by-column). This dimensional approach to processing allows the system to maintain data continuity in storage while systematically addressing the access inefficiency through structured phase-based processing.
Data Source
AI summary
A column reader, a one-dimensional transformer, and a row writer repeat reading of one or more columns, one-dimensional transform of one or more columns, and writing of one or more rows until all original data of M rows and N columns is transformed and recorded in an auxiliary storage device as intermediate data of N rows and M columns. A column reader, a one-dimensional transformer, and a row writer repeat reading of one or more columns, one-dimensional transform of one or more columns, and writing of one or more rows until all of the intermediate data is transformed and recorded in the auxiliary storage device as final data of M rows and N columns.