Research on Jukes-Cantor Model Parallel Algorithm Based on OpenMP

The parallel algorithm of the evolutionary distance between different species is implemented by using OpenMP parallel technique in this paper. In order to get the best degree of algorithm parallelism, the method of making loop variable corresponding to the rower and column labels of distance matrix is adopted. It is to say that the double loop can be converted into single loop to improve the parallel efficiency. The serial algorithm and parallel algorithm are compared in this paper. The experiment result shows that the highest speedup is 14.1. It improves the running efficiency of the program. It is a great significance to dealing with massive bioinformatics data.