Application of high performance computing techniques to the semantic data transformation

Abstract

The growth of the Life Science Semantic Web is illustrated by the increasing number of resources available in the Linked Open Data Cloud. Our SWIT tool supports the generation of semantic repositories, and it has been successfully applied in the field of orthology resources, helping to achieve objectives of the Quest for Orthologs consortium. However, our experience with SWIT reveals that despite the computational complexity of the algorithm is linear with the size of the dataset, the time required for the generation of the datasets is longer than desired. The goal of this work is the application of High Performance Computing techniques to speed up the generation of semantic datasets using SWIT. For this purpose, the SWIT kernel was reimplemented, its algorithm was adapted for facilitating the application of parallelization techniques, which were finally designed and implemented. An experimental analysis of the speed up of the transformation process has been performed using the orthologs database InParanoid, which provides many files of orthology relations between pairs of species. The results show that we have been able to obtain accelerations up to 7000x. The performance of SWIT has been highly improved, which will certainly increase its usefulness for creating large semantic datasets and show that HPC techniques should play an important role for increasing the performance of semantic tools.