A spatial-temporal data management structure for big internet of things data

In recent year, the concept of the Internet of Things (IoT) has been attracting attention from various fields, as IoT devices can continuously monitor various environmental properties. While the number of IoT devices increases rapidly, managing large volume of IoT data faces a challenging scalability issue. To address this scalability issue, many studies have shown that the performance of key-value storages is better than traditional relational databases. However, IoT data have multi-dimensional attributes including spatial, temporal and thematic attributes. How to construct an efficient multi-dimensional index structure based on a key-value storage has become a popular topic in recent years. In this research, we consider four main types of attribute/query: spatial, temporal, keyword, and value. While each type of the attribute has its own suitable indexing method, integrating the indexing methods usually requires a certain sequence. However, this sequence of indexing structure is one of the key factors deciding the query performance. While many literatures directly present their designed sequence, this research proposes an adaptive method to decide the indexing sequence based on query criteria and the selectivity and performance of different indexing methods. In principle, highly selective queries should be performed first to reduce the number of intermediate results, which could improve the query performance of following queries. Based on this idea, this research proposes an indexing structure considering every possible sequence and automatically identifying the most efficient one for each individual query. As a result, the proposed solution can significantly improve the query performance of multi-dimensional IoT sensor data.