Content-aware data distribution over cluster nodes

Proper data items distribution may seriously improve the performance of processing in distributed environment. However, typical datastorage systems as well computational frameworks do not pay special attention to that aspect. In this paper author introduces two custom addressing methods for on example Scalable Distributed Two-Layer Datastore. The basic idea those is preserve stored same cluster node are similar each other following concepts clustering. Still, most clustering mechanisms have serious problem with scalability which a severe limitation Big Data applications. proposed allow efficiently distribute set over buckets. As it was shown by experimental results, all generate good results comparison traditional techniques like k-means, agglomerative and birch environment experiments proper can effectiveness processing.