Optical Network Optimization via Parallel Lightpath Sequence Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing optical networks, such as routing and wavelength assignment, often fail to guarantee optimal results due to the variability in lightpath demand sequences, leading to suboptimal performance.
Innovation Solution
Generating multiple lightpath demand sequences with different orders and processing them in parallel using a distributed system like Hadoop on a cloud computing platform to compare results and obtain the optimum outcome.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single lightpath demand sequence is processed using heuristic algorithm, then the computation is simple and fast, but the optimization result is suboptimal and not guaranteed to be the best
Solution Approach 1:
The patent segments the optimization problem by dividing the lightpath demand set into multiple different sequences (e.g., through random shuffling or different sorting criteria). Each sequence is processed independently by separate processing terminals using the same heuristic algorithm, thereby transforming a single complex optimization challenge into multiple simpler sub-problems that can be solved in parallel.
Solution Approach 2:
The patent merges the results from multiple parallel processing terminals by comparing the optimization outcomes (such as wavelength usage, routing efficiency, or number of lightpaths) and selecting the best result. This combination approach ensures that the final optimization result is the best among all possibilities, significantly improving the reliability of the optimization while distributing the computational complexity across multiple terminals.
2Reliability
If multiple lightpath demand sequences are processed to ensure optimal result, then the optimization quality improves, but the computation time increases
Solution Approach 1:
The patent divides the computation task by creating multiple lightpath demand sequences and assigning each sequence to a separate processing terminal. This segmentation allows simultaneous processing of multiple sequences, effectively parallelizing the computation and reducing the total time required compared to sequential processing.
Solution Approach 2:
The patent performs preliminary actions by pre-generating multiple lightpath demand sequences with different orders before the actual optimization computation. These pre-prepared sequences are then ready for parallel processing, avoiding the need to generate and process sequences sequentially during the optimization phase, thus reducing overall computation time.
3Productivity
If parallel processing is used to obtain multiple results, then the computation efficiency improves, but the system complexity increases
Solution Approach 1:
The patent employs multiple processing terminals that all execute the same heuristic optimization algorithm, making each terminal universal in its function. Each terminal can independently process any lightpath demand sequence with the same optimization logic, simplifying the system design compared to having specialized different algorithms for different terminals, while still achieving parallel processing efficiency.
Solution Approach 2:
The patent creates copies of the processing terminals, where each terminal is an identical copy capable of performing the same optimization task. This copying approach simplifies the system architecture by using replicated identical units rather than complex differentiated components, making the parallel processing system easier to implement and manage while maintaining high computation efficiency.
Data Source
AI summary
A method for optimizing an optical network includes: obtaining a lightpath demand set including multiple lightpath demands; generating multiple lightpath demand sequences with different orders from the lightpath demand set; obtaining results of the respective lightpath demand sequences in parallel by multiple processing terminals; obtaining all the results, and comparing the results to obtain an optimum result. In this embodiment, by generating multiple lightpath demand sequences with different orders for the lightpath demand set and obtaining many results, the optimum result can be selected out from the results. Furthermore, multiple processing terminals may obtain the result in parallel, thus improving an operation efficiency.