Multi-Core GPU Automation Controller Parallel Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automation systems in industrial applications rely on CPUs for control tasks, which are limited by the number of available parallel execution units, requiring complex real-time systems and task decomposition to manage multiple tasks simultaneously, and GPUs are primarily used for graphic applications, not automation.
Innovation Solution
Utilizing multi-core graphics processing units (GPUs) to execute control tasks and algorithms in automation systems, leveraging their parallel architecture to simplify real-time execution and enhance performance by running tasks on dedicated calculation units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If CPUs are used for control tasks in automation systems, then reliability and real-time performance are maintained, but processing power and parallel execution capability are limited
Solution Approach 1:
The patent applies multi-functionality by enabling GPUs, originally designed for graphic rendering, to execute control tasks in automation systems. The system allows a single GPU to perform both traditional graphic operations and industrial control functions, eliminating the need for separate dedicated hardware for each function and thereby increasing processing power while managing system complexity.
Solution Approach 2:
The patent substitutes the traditional CPU-based control system with a GPU-based parallel processing architecture. By replacing the sequential processing mechanism of CPUs with the parallel processing capability of GPUs, the system achieves significantly higher processing power for control tasks while maintaining real-time performance through hardware-level parallelism.
2Productivity
If more automation tasks are executed than CPUs are available, then task completion is achieved, but complex real-time systems and task decomposition are required
Solution Approach 1:
The patent applies segmentation by dividing control tasks into multiple independent parallel threads that can be executed simultaneously on different GPU cores. This task decomposition is simplified compared to traditional real-time systems because the GPU architecture naturally supports parallel thread execution, reducing the complexity of real-time scheduling and synchronization mechanisms.
Solution Approach 2:
The patent transitions from sequential task execution on a single CPU to parallel task execution across multiple GPU cores, adding a dimensional aspect of parallelism. This dimensional change from one-dimensional sequential processing to multi-dimensional parallel processing enables simultaneous execution of multiple automation tasks without requiring complex real-time operating systems.
3Adaptability or versatility
If GPUs are used for graphic applications only, then graphic rendering performance is optimized, but processing power is underutilized for automation tasks
Solution Approach 1:
The patent implements multi-functionality by designing a system where GPUs serve dual purposes: traditional graphic rendering and industrial automation control tasks. This universal approach allows the same hardware resource to be shared between different application domains, increasing adaptability while fully utilizing the available processing power through time-multiplexed or space-multiplexed execution.
Data Source
AI summary
A method and system are provided for performing the computational execution of automation tasks with automation devices by combining one or more central processing units (CPU) and one or more Graphics Processing Units (GPU). The control tasks and/or control algorithms are executed by the single-core or multi-core control unit (CPU) and a multi-core-graphics processor (GPU) or both in parallel at the same time.