CGRA Mapping Verification Using Data Flow Graph Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge lies in verifying the accuracy of mapping results produced by automatic mappers for coarse-grained reconfigurable architectures (CGRA), as manual verification is impractical and existing methods fail to ensure correctness and improve mapping accuracy.
Innovation Solution
A non-transitory computer-readable recording medium and apparatus that includes a mapping result verification program to verify the accuracy of mapping results by comparing the output of automatic mappers with a data flow graph, using symbolic execution and verification units to check the correctness of arithmetic operations and data flow in CGRA.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic mapping is used to map data flow graph to CGRA, then mapping speed and productivity are improved, but mapping accuracy and reliability deteriorate due to potential bugs in the automatic mapper
Solution Approach 1:
The patent implements a verification mechanism that automatically checks mapping results by comparing the mapped CGRA configuration against the original data flow graph. This feedback loop detects mapping errors and alerts users, thereby maintaining high productivity while improving reliability through automated error detection.
Solution Approach 2:
The patent introduces an intermediary verification module that acts as a mediator between the automatic mapper and the final mapping result. This intermediary component validates the mapping without requiring manual intervention, thus preserving productivity while enhancing reliability through an additional verification layer.
2Reliability
If manual mapping is performed to ensure accuracy, then mapping reliability is improved, but mapping time and productivity deteriorate
Solution Approach 1:
The patent creates a verification copy of the mapping result and compares it against the original data flow graph. This copying mechanism allows automated verification of mapping accuracy without requiring manual mapping, thus maintaining high productivity while ensuring reliability through comparison validation.
3Reliability
If verification processes are added to check mapping results, then mapping reliability is improved, but system complexity and device complexity increase
Solution Approach 1:
The verification system performs self-service by automatically comparing the mapped configuration against the original data flow graph without requiring external manual verification. This self-verification mechanism improves reliability while minimizing additional complexity by using the existing mapping components for validation.
Data Source
AI summary
A non-transitory computer-readable recording medium stores therein a mapping result verification program that causes a computer to execute a process including acquiring both of a data flow graph that represents predetermined calculation including a plurality of arithmetic operations and a mapping result obtained by mapping the data flow graph into a CGRA that includes a plurality of arithmetic operation units, and verifying, regarding each of first arithmetic operation units to which the arithmetic operations have been respectively allocated in the mapping result, based on an arithmetic operation result that has been obtained by performing the arithmetic operations, whether or not the mapping result matches the data flow graph.