Parallel Sorting by Approximate Splitting for Multi-core Processors

Many pivot-based sorting methods benefit from balanced partition of data into smaller chunks with better parallelism and locality. However, it is difficult to partition data with the same high efficiency. This paper presents an improved partition method, Parallel Sorting by Approximate Splitting, which is based on an extend pivots selecting algorithm. Compared with other algorithm, such as PSRS, its splitter method is more flexibility and efficiency. Experiments which implemented by OpenMP in multi-cores environment and the theoretical analysis show the method is an efficiency algorithm with very good scalability.