A new exact algorithm for the shortest path problem: An optimized shortest distance matrix

摘要

The growing amount of data generated from the increasingly sophisticated network connections requires greater accuracy and higher efficiency in pinpointing the shortest paths concerned. Therefore, the earlier classical exact algorithms are no longer 100 percent suitable for large-scale data processing, for their known great time complexity during calculation. In this paper, We present an updated shortest distance matrix (SDM) algorithm. Evidence to the operation's properties is provided and the properties are used in subsequent optimizations. Compared with Dijkstra's algorithm and Floyd's algorithm, the optimized SDM algorithm with parallel mode makes a great improvement in shortening the running time. The data test shows that the new algorithm improves the efficiency in processing a large amount of data.