Robust Massively Parallel Sorting

Michael Axtmann, Peter Sanders
2017 2017 Proceedings of the Ninteenth Workshop on Algorithm Engineering and Experiments (ALENEX)  
We investigate distributed memory parallel sorting algorithms that scale to the largest available machines and are robust with respect to input size and distribution of the input elements. The main outcome is that three sorting algorithms cover the entire range of possible input sizes. For all three algorithms we devise new low overhead mechanisms to make them robust with respect to duplicate keys. The one for medium sized inputs is a new variant of quicksort with fast high-quality pivot
more » ... on. Asymptotic analysis at the same time provides performance guarantees and guides the selection and configuration of the algorithms. We validate these hypotheses using extensive experiments on 7 algorithms, 10 input distributions, up to 262 144 cores, and varying the input sizes over 9 orders of magnitude. For "difficult" input distributions, our algorithms are the only ones working at all. For all but the largest input sizes, we are the first to perform experiments on such large machines at all and our algorithms significantly outperform the ones on would conventionally have considered.
doi:10.1137/1.9781611974768.7 dblp:conf/alenex/Axtmann017 fatcat:npylsyoyzzblxdltie5ze5phoq