RadixSort

C implementation of merge sort, radix sort and parallel radix sort on 32 bit unsigned integers

Performance of qsort and merge sort degrades as data becomes more unordered, while running time of radix sort is much less affected.

Parallel radix sort is only worth it if the number of elements is large.

Tested on an i5-4210U with 2 cores. Algorithm will scale gracefully with higher number of cores.

For very small arrays, qsort, merge sort and sorting networks should be favoured over radix sort.