Computational Memory Zero Disable With Matrix Error Checking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deep learning neural networks require significant power consumption due to data movement between memory and processing elements, leading to inefficiencies such as increased complexity, processing time, and larger chip area requirements, particularly in battery-powered devices.
Innovation Solution
A processing device with an array of processing elements connected for direct communication, featuring an input zero detector and zero disable circuit to optimize data movement, and a controller that adds error-checking values to matrices for efficient parallel operations and error detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data is moved between memory and processing elements in traditional computer architecture, then computations can be performed, but power consumption increases significantly
Solution Approach 1:
The system segments the computational array into multiple independently controllable processing elements that can be selectively activated. Each processing element can be enabled or disabled based on whether its input data is non-zero, allowing fine-grained power management while maintaining computational throughput across the array.
Solution Approach 2:
The system dynamically adjusts the operational state of processing elements based on input data characteristics. The enablement/disablement of processing elements is determined in real-time based on whether input values are zero or non-zero, creating a dynamic power management system that adapts to the computational workload.
2Use of energy by moving object
If processing elements are disabled to save power, then energy consumption decreases, but processing capability is reduced
Solution Approach 1:
The system applies different operational states to different processing elements based on their specific input data. Rather than uniformly disabling entire blocks of processing elements, the system selectively disables only those individual elements receiving zero inputs, maintaining processing capability in regions with non-zero data while saving power in regions with zero data.
3Reliability
If zero inputs are processed normally, then computational accuracy is maintained, but unnecessary power is consumed
Solution Approach 1:
The system converts the potentially harmful effect of zero inputs (which waste power when processed) into a beneficial control signal. The zero detector identifies zero inputs and uses them to trigger the disablement of processing elements, thereby transforming what would be a power-wasting condition into a power-saving opportunity while maintaining computational accuracy through proper handling of zero values.
4Reliability
If error checking is implemented in neural network computations, then computational reliability improves, but processing complexity increases
Solution Approach 1:
The system performs preliminary error detection by monitoring input values for zero conditions before they enter the main computational path. By detecting zeros upfront and disabling affected processing elements, the system prevents erroneous computations from propagating through the array, providing a form of preliminary error checking that simplifies the overall error handling mechanism.
Data Source
AI summary
A processing element includes an input zero detector to detect whether the input from the neighbor processing element contains a zero. When the input from the neighbor processing element contains the zero, a zero disable circuit controls the input from the neighbor processing element and respective data of the memory to both appear as unchanged to the arithmetic logic unit for the operation. A controller of an array of processing elements adds a row of error-checking values to a matrix of coefficients, each error-checking value of the row of error-checking values being a negative sum of a respective column of the matrix of coefficients. The controller controls a processing element to perform an operation with the matrix of coefficients and an input vector to accumulate a result vector. Owing to the error-checking values, when a sum of elements of the result vector is non-zero, an error is detected.