Conscious Radix Sort

The exploitation of data locality in parallel computers is paramount to reduce the memory traac and communication among processing nodes. We focus on the exploitation of locality by Parallel Radix sort. The original Parallel Radix sort has several communication steps in which one sorting key may have to visit several processing nodes. In response to this, we propose a reorganization of Radix sort that leads to a highly local version of the algorithm at a very low cost. As a key feature, our algorithm performs one only communication step, forcing keys to move at most once between processing nodes. Also, our algorithm reduces the amount of data communicated. Finally , the new algorithm achieves a good load balance which makes it insensitive to skewed data distributions. We call the new version of Parallel Radix sort that combines locality and load balance, Communication and Cache Conscious Radix sort (C 3-Radix sort). Our results on 16 processors of the SGI O2000 show that C 3-Radix sort reduces the execution time of the previous fastest version of Parallel Radix sort by 3 times for data sets larger than 8M keys and by almost 2 times for smaller data sets. 1 Introduction Large scale parallel computers include complex memory hierarchies and communication networks. With the design of software that exploits data locality, both the memory hierarchy and the communication network beneet from lower data traac. We address the important problem of locality exploitation for Parallel Radix sort. Radix sort is the most compet