Adaptive Multiplier for Mixed-Format Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern processor chips face high resource consumption due to the need for multiple multipliers to handle various numerical operations in different formats, leading to increased chip area usage and resource allocation.
Innovation Solution
A data processing method that involves acquiring and converting input data formats to a unified target format, using an adaptive multiplier to perform operations, and determining truncation bit widths to obtain a multiplication result, thereby reducing resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple multipliers are designed for different calculation requirements and data formats, then the processor can handle various numerical operations, but the chip area consumption increases significantly
Solution Approach 1:
The patent implements a universal multiplier that can handle multiple data formats (integer, floating-point, fixed-point) through a unified architecture. The multiplier uses format detection circuits to identify input data types and automatically configures itself to perform the appropriate multiplication operation, eliminating the need for separate dedicated multipliers for each data format. This multi-functional approach allows a single multiplier to replace what would traditionally require multiple format-specific multipliers, thereby reducing chip area while maintaining comprehensive calculation capability.
Solution Approach 2:
The patent employs parameter change by dynamically adjusting the multiplier's internal configuration based on the detected data format. The system changes operational parameters such as bit-width allocation, normalization factors, and result formatting according to the input data types. This allows the same hardware structure to adapt its behavior to different calculation requirements, achieving versatility without requiring multiple fixed-architecture multipliers.
2Reliability
If different multipliers are designed for different data formats, then each format can be processed accurately, but device resource consumption increases
Solution Approach 1:
The universal multiplier maintains calculation accuracy across different data formats by incorporating format-specific processing logic within a single hardware structure. The multiplier detects the input format and applies the appropriate calculation methodology, ensuring precise results for integers, floating-point numbers, and fixed-point values alike. This approach preserves the reliability benefits of format-specific multipliers while avoiding the resource multiplication that would result from implementing separate hardware units for each format.
Solution Approach 2:
The patent combines multiple format-specific processing capabilities into a single integrated multiplier unit. By merging the functional capabilities of what would traditionally be separate integer multipliers, floating-point multipliers, and fixed-point multipliers into one unified structure, the system achieves the same computational reliability with significantly reduced device resource consumption.
3Quantity of substance
If a single multiplier is used for multiple data formats, then device resource consumption is reduced, but the device complexity increases
Solution Approach 1:
The universal multiplier is segmented into distinct functional modules: a format detection unit that identifies input data types, a configuration unit that sets operational parameters based on the detected format, and a computation unit that performs the actual multiplication. This modular segmentation organizes the complexity into manageable, independent components that can be systematically controlled, making the overall system easier to manage despite handling multiple formats.
Solution Approach 2:
The multiplier employs dynamic configuration where the operational parameters and internal structure are adjusted in real-time based on the detected data format. The system transitions between different operational states (integer mode, floating-point mode, fixed-point mode) dynamically, allowing a single structure to adapt its behavior rather than requiring multiple static structures. This dynamic adaptability reduces device resources while the complexity is managed through automated format detection and configuration.
Data Source
AI summary
The present application relates to a data processing method, a computer device, and a storage medium. The method includes: acquiring data formats of two pieces of input data; the data formats of the two pieces of input data being the same; determining a target data conversion algorithm matching the data formats from a plurality of preset data conversion algorithms, and performing, by using the target data conversion algorithm, data format conversion on the two pieces of input data to obtain at least two pieces of target input data; processing, by using a multiplier, the at least two pieces of target input data to obtain a preliminary operation result; and determining truncation bit widths corresponding to the two pieces of input data, and processing the preliminary operation result according to the truncation bit widths, to obtain a multiplication operation result corresponding to the two pieces of input data.