Group Theoretic Tabu Search for GPU Parallel Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optimization methods, such as exact methods and heuristic approaches, face challenges in efficiently solving complex optimization problems like the n-city, m-Traveling Salesman Problem within time constraints, often resulting in suboptimal solutions or high computational stress.
Innovation Solution
The implementation of group theoretic tabu search (GTTS) processing, which fragments the solution space and distributes partitions across multiple GPU threads for parallel processing, allowing for timely and high-quality solution determination by iteratively refining fragmentations until stopping criteria are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If exact methods or heuristic approaches are used to solve complex optimization problems, then solution quality can be maintained, but computational time and resource requirements become excessively high
Solution Approach 1:
The patent divides the solution space into multiple partitions or fragments, assigning each partition to a separate GPU thread for parallel processing. This segmentation allows the optimization problem to be solved simultaneously across multiple processing units, dramatically reducing computational time while maintaining solution quality through comprehensive exploration of the divided solution space
2Speed
If traditional sequential processing methods are used, then implementation complexity is low, but processing speed and optimization efficiency are insufficient
Solution Approach 1:
The patent replaces traditional CPU-based sequential processing with GPU-based parallel processing architecture. This substitution leverages the massively parallel computational capability of GPUs, which contain thousands of smaller cores optimized for simultaneous execution of multiple threads, thereby achieving significant speedup for optimization problems at the cost of increased architectural complexity
3Reliability
If the solution space is explored thoroughly to ensure optimal solutions, then solution reliability improves, but computational resources and time consumption increase
Solution Approach 1:
By partitioning the solution space into multiple fragments and processing them in parallel across GPU threads, the system achieves thorough exploration of the solution space for reliable results while distributing computational energy consumption across multiple efficient processing units, reducing the energy burden on any single processor
Data Source
AI summary
Systems, devices, methods, and computer-readable media are disclosed for utilizing group theoretic techniques to enable data exchange between a supervisory central processing unit (CPU) and a group of graphical processing units (GPUs). The CPU may be configured to utilize a tabu search metaheuristic to explore a solution space to determine an optimal solution to an optimization problem. More specifically, the CPU may determine a fragmentation of a solution space that yields multiple partitions of the solution space and may assign each partition to a respective GPU configured to calculate a computational result. The CPU may then determine a new fragmentation of the solution space based on the computational results received from the GPUs that yields new partitions of the solution space and may assign each new partition to a respective GPU configured to again generate a computational result based on its assigned new partition. The CPU may continue to determine new fragmentations based on the computational results of the GPUs until stopping criteria are satisfied and a timely, high-quality solution to the optimization problem is determined.