SNPs are fundamental roles for various applications including medical diagnostic, phylogenies and drug design. They provide the highest-resolution genetic fingerprint for identifying disease associations and human features. Genetic variants that are near each other tend to be inherited together; these regions of linked variants are known as haplotypes. Recently, genetics researches revealed that SNPs within certain haplotype blocks induce only a few distinct common haplotypes in the majority of the population. The existence of haplotype block structure has serious implications for association-based methods for the mapping of disease genes. This paper proposes a parallel haplotype block partition and SNPs selection method under a diversity function by using the Hadoop MapReduce framework. The experiment shows that the proposed MapReduce-paralleled combinatorial algorithm performs well on the real-world data obtained in from the HapMap data set; the computation efficiency can be significantly improved proportional to the number of processors being used.