ENOD sort on diamond architecture
Kamal Jadidy Aval, Masumeh Damrudi · International Conference on Software Engineering · 2010
In this paper, we are extending The NOD sorting algorithm which is implemented on Diamond architecture. This algorithm which we named it ENOD (Extended Neighborhood sort On Diamond) sorts n data elements with 7/4n processors. However, most popular environments provide little explicit support for parallelism, leading to the common view that concurrency is hard. There are architectures and sorting algorithms that are used but we always endeavor to find new and optimal ones. The algorithm on the Diamond architecture sorts data elements using 7/4n processors with a running time of O(log2 n). This architecture is heterogeneous and uses cheaper processors more than expensive ones. Though, this architecture with the issued algorithm makes a tradeoff between number of processors and their cost. The ENOD algorithm is simpler and more intuitive than the algorithms that are available.