Graphical Model Pattern Recognition for Concurrent Multi-Core Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex engineering and scientific computations often require significant computational resources, as single processors can be limiting in handling large and complex models, leading to time-consuming simulations, especially when multiple parameter variations need to be analyzed.
Innovation Solution
The method involves recognizing patterns in graphical models suitable for concurrent processing and deploying them on multi-threaded, multi-core, or multi-processor systems, allowing for simultaneous execution of computations across multiple resources, such as scatter and gather, pipelining, and iteration patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If single processor computing is used to execute graphical models, then device complexity is reduced, but computation time increases significantly for large and complex models
Solution Approach 1:
The patent segments the graphical model execution into multiple independent partitions that can be processed concurrently. The pattern recognizer identifies suitable patterns in the graphical model, and the pattern partitioner divides these patterns into separate partitions that can be executed in parallel on multi-core or multi-processor systems, thereby reducing overall computation time while distributing the computational load across multiple resources.
Solution Approach 2:
The patent transitions from single-dimensional sequential processing to multi-dimensional parallel processing by utilizing multiple computing cores or processors simultaneously. This dimensional expansion allows the system to execute multiple partitions of the graphical model concurrently, effectively reducing computation time by leveraging additional computational dimensions rather than relying solely on sequential execution.
2Adaptability or versatility
If multiple parameter variations are analyzed in complex models, then analysis completeness is improved, but computation time increases significantly
Solution Approach 1:
The patent applies segmentation by dividing the analysis of multiple parameter variations into separate executable partitions. Each partition can handle specific parameter variations independently, allowing the system to process multiple parameter sets concurrently rather than sequentially. This segmentation enables comprehensive parameter analysis while maintaining reduced computation time through parallel execution on multi-core systems.
3Productivity
If conventional sequential processing is used for graphical models, then ease of operation is maintained, but productivity decreases for large-scale computations
Solution Approach 1:
The patent implements self-service through automated pattern recognition and partitioning. The system automatically identifies suitable patterns in the graphical model and partitions them for parallel execution without requiring manual intervention. This automation maintains ease of operation by hiding the complexity of parallel processing from the user, while simultaneously improving productivity through efficient concurrent execution of model partitions.
Data Source
AI summary
In one embodiment, a computer-implemented method for concurrently processing at least a portion of a graphical model is provided. The method may include obtaining the graphical model; recognizing a pattern in the graphical model, the pattern suitable for concurrent processing; and employing concurrent processing using multi-thread, multi-core, or multi-processor computing device when executing the pattern in the graphical model.