Super Scalar Sample Sort

Abstract. Sample sort, a generalization of quicksort that partitions the input into many pieces, is known as the best practical comparison based sorting algorithm for distributed memory parallel computers. We show that sample sort is also useful on a single processor. The main algorithmic insight is that element comparisons can be decoupled from expensive conditional branching using predicated instructions. This transformation facilitates optimizations like loop unrolling and software pipelining. The final implementation, albeit cache efficient, is limited by a linear number of memory accesses rather than the O(n log n) comparisons. On an Itanium 2 machine, we obtain a speedup of up to 2 over std::sort from the GCC STL library, which is known as one of the fastest available quicksort implementations. 1 Introduction Counting comparisons is the most common way to compare the complexity ofcomparison based sorting algorithms. Indeed, algorithms like quicksort with good sampling strategies [9, 14] or merge sort perform close to the lower bound oflog n! ss n log n comparisons3 for sorting n elements and are among the bestalgorithms in practice (e.g., [24, 20, 5]). At least for numerical keys it is a bi