Indexing spatial data in cloud data managements

With the proliferation of smart phones and location-based services, the amount of data with spatial information, referred to as spatial data, is dramatically increasing. Cloud computing plays an important role handling large-scale data analysis, and several cloud data managements (CDMs) have been developed for processing data. CDMs usually provide key-value storage, where each key is used to access its corresponding value. However, user-generated spatial data are usually distributed non-uniformly. In this paper, we present a novel key design based on an R⁺-tree (KR⁺-index) for retrieving skewed spatial data efficiently. In the experiments, we implement the KR⁺-index on Cassandra, and study its performance using spatial data. Experiments show that the KR⁺-index outperforms the-state-of-the-art methods.