QCCF-tree: A New Efficient IoT Big Data Indexing Method at the Fog-Cloud Computing Level

Over the past decades, the volume of Internet of Things (IoT) data has exploded which raise the problem of indexing, storing and retrieving efficiently. Most studies are based on dividing the target dataset into subsets using balls with one or two pivots. However, in the era of big data, where efficient indexing is important, the subspace volumes grow exponentially, which could degenerate the index. This problem is due to the inherent inadequacy of space partitioning. The topology must avoid biased allocation of objects for separable sets and must not influence the index structure. To meet this criteria, in this paper a new indexing structure called QCCF-tree (Quad tree based on Containers at the Cloud- Fog computing level), is proposed, based on dividing the space into four balls with four pivots. For the enhancement of the retrieving time in this new structure, the query is searched in parallel at each node of the QCCF-tree. The experimental evaluation of the proposed system, compared with several indexing systems, show that QCCF-tree outperforms most indexing systems either in the construction or the similarity query search.