A fast estimation of shortest path distance for power-law network predominant cloud service
Jin Liu, Chunming Rong, Ganseng Zhao
Abstract
Jin Liu, Chunming Rong, Ganseng Zhao
Abstract
The estimation of the shortest path between two vertices in a network graphs is an important issue for many applications in the real world, which may be road networks, social collaboration networks, biological networks and so forth. The short response time, the less space cost and the high accuracy are three critical evaluation metrics in the approximate calculation of the shortest path. To achieve a quick and accurate calculation of an approximate shortest path distance, the typical method that takes a non-trivial approach calculates the distance between every pair of vertices in advance and records the calculation results with an n×n matrix, where n is the number of vertexes in the network graph. Unfortunately, it hard for this method function because it is difficult to sustain the huge amount of storage space and time consuming preprocessing for the big size of the network graph in practice. Unlike many efforts that have been made to minimize the costs of time and space in enhancing the accuracy of the shortest path calculation, this paper proposes an estimation scheme of the shortest path calculation for power-law graphs, due to the fact that many networks in the real applications are power-law networks. The theoretical analysis indicates that the schema is successfully optimized so that the query time can be an approximation constant and the calculation result of the approximate shortest path distance is almost 2 to 3 times as long as the actual shortest path distance between a pair of vertices. A simulation experiment also validates the analysis and demonstrates the feasibility of the proposed scheme.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The estimation of the shortest path between two vertices in a network graphs is an important issue for many applications in the real world, which may be road networks, social collaboration networks, biological networks and so forth. The short response time, the less space cost and the high accuracy are three critical evaluation metrics in the approximate calculation of the shortest path. To achieve a quick and accurate calculation of an approximate shortest path distance, the typical method that takes a non-trivial approach calculates the distance between every pair of vertices in advance and records the calculation results with an n×n matrix, where n is the number of vertexes in the network graph. Unfortunately, it hard for this method function because it is difficult to sustain the huge amount of storage space and time consuming preprocessing for the big size of the network graph in practice. Unlike many efforts that have been made to minimize the costs of time and space in enhancing the accuracy of the shortest path calculation, this paper proposes an estimation scheme of the shortest path calculation for power-law graphs, due to the fact that many networks in the real applications are power-law networks. The theoretical analysis indicates that the schema is successfully optimized so that the query time can be an approximation constant and the calculation result of the approximate shortest path distance is almost 2 to 3 times as long as the actual shortest path distance between a pair of vertices. A simulation experiment also validates the analysis and demonstrates the feasibility of the proposed scheme.
Key concepts: Shortest path problem, Distance, Average path length, Computer science, Constrained Shortest Path First, Shortest Path Faster Algorithm, K shortest path routing, Euclidean shortest path