A parallelism-motivated sequential sorting framework

We employ techniques developed and used in the design of parallel sorting algorithms to propose a new framework for sequential sorting. This framework is then used to design new deterministic and randomized sorting methods whose asymptotic worst-case running time can match the existing lower bound for sorting, yet their practical performance, as witnessed by an experimental study, surpasses existing optimized sorting algorithm implementations. We adapt in the proposed new framework techniques used for parallel sorting such as deterministic regular sampling and random oversampling. We extend the notion of deterministic regular sampling into deterministic regular oversampling for sequential sorting and show its potential along with the previously available technique of random oversampling. We then show how our newly developed techniques can utilize and potentially speedup several existing sequential sorting algorithms. Experimental results based on an implementation of our two methods support our efficiency claims. Our new approach maintains better locality of reference and can naturally benefit from multicore architectures. This is to our knowledge the first time that sequential computing can beneficially draw from parallel computing techniques.