Scheduling algorithms for the revised pdpsort

Parallel sorting algorithms for manycore CPU systems are needed in data science and big data era. The PDPSort (Parallel Dual Pivot STLSort) can be revised and extended to achieve higher and more stable Speedups. This paper experiments four scheduling algorithms, Always Left (LAL), Always Right (RAL), Longer Partition First (LPF) and Shorter Partition First (SPF). Eventually, the revised PDPSort can achieve faster Speedup by upto 7.35×, 4.83× and 4.43× over the STLSort on AMD R7-1700, AMD FX-8320, and Intel i7-2600 Linux systems, respectively. Moreover, the LPF algorithm yields more stable Speedup's than others.