Multi-Core Vehicle Control Device Data Coherence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for operating control devices in vehicles, particularly multi-core processors, face challenges in providing reliable and efficient functional validation due to complexities in data coherence and communication bandwidth, especially when dealing with distributed calculation processes across multiple cores.
Innovation Solution
A method where computing processes are assigned predetermined system cycle times, and output signals are provided only after all processes with the same cycle time are completed, allowing for temporal coherence and reducing communication bandwidth by not revealing core architecture or calculation distribution, with optional start signals triggered by system events or states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If output signals are provided after each core completes its calculations independently, then communication bandwidth is reduced, but data coherence and temporal consistency deteriorate
Solution Approach 1:
The patent merges the output signals from multiple cores by collecting all intermediate and final values from parallel computing processes into a unified data structure. This unified structure is then processed together to generate a single coherent output signal, ensuring data consistency while maintaining the bandwidth efficiency of parallel processing.
Solution Approach 2:
The patent introduces an intermediary buffering mechanism that temporarily stores intermediate and final values from multiple cores. This buffer acts as a mediator that synchronizes data from different cores before generating the output signal, ensuring temporal coherence without requiring continuous high-bandwidth communication between cores.
2Loss of information
If the output signal reflects the core architecture and calculation distribution, then diagnostic information is improved, but system complexity and communication overhead increase
Solution Approach 1:
The patent extracts only the necessary diagnostic information (intermediate and final values) from the computing processes while leaving out the complex architectural details. The output signal contains the extracted functional data without embedding information about core distribution or architectural configuration, simplifying communication while preserving diagnostic capability.
Solution Approach 2:
The patent creates a simplified copy of the computational data that represents the system's functional state without replicating the complex architectural structure. This copy contains the essential diagnostic values in a standardized format that can be transmitted efficiently without revealing implementation-specific architectural details.
3Productivity
If computing processes are assigned to different cores for parallel execution, then productivity is improved, but ensuring all processes complete before output increases coordination complexity
Solution Approach 1:
The patent performs preliminary assignment of computing processes to specific cores based on their characteristics and dependencies. This pre-planning allows the system to execute multiple processes in parallel while minimizing the coordination required during runtime, as the distribution strategy is established before execution begins.
Solution Approach 2:
The patent implements a feedback mechanism where the control device monitors the completion status of computing processes across multiple cores. Based on this feedback, the system dynamically determines when all necessary calculations are complete and proceeds to generate the output signal, ensuring coordination without requiring complex real-time synchronization protocols.
Data Source
Figure 1~2B
Figure 3
AI summary
The invention relates to a method for operating a control device for a vehicle. The control device is designed as a multi-core processor having at least two cores for the parallel execution of computational processes. The control device comprises a communications interface, via which same can be coupled to a diagnosis unit via signals. In the method, the computational processes are each assigned to a predetermined system cycle duration of the control device. When all computational processes that are assigned to the same predetermined system cycle duration are completed, an output signal is then provided via the communications interface of the control device. The output signal comprises at least one of the input-, output- or intermediate values of the computational processes. The invention also relates to a diagnosis system.