ParallelBucketSort is a final project of the Parallel and Distributed Computing course, whose goal was to design and implement a parallel and high-scalable version of the [Bucket sort] (https://en.wikipedia.org/wiki/Bucket_sort) sorting algorithm by using Skandium.
Additionally, in the current repository you can find the laboratory report (italian only).
The project can be compiled by typing:
ant
In order to generate the input files (random floating-points array), you have to type:
ant RandomDoubleArrays -Dn=nValue -Dm=mValue
where nValue and mValue are the size of each array and the number of arrays you want to generate, respectively.
The generated array can be sorted by typing:
ant BucketSortAlgorithm -Dnb=nBuckets -Dpg=parDegree
where BucketSortAlgorithm can be:
SequentialBucketSort, for running the sequential version of the BucketSort algorithm;BucketSortMap, for running the parallel version of the BucketSort algorithm;
and nBuckets is the number of buckets, parDegree is the parallel degree.
In order to check the correct sorting of the input arrays, you can type:
ant AreSorted