Metadata-guided Evaluation of Resource-constrained Queries in Content Caching Based Wireless Networks

Recent years have witnessed the emergence of data-centric storage that provides energy-efficient data dissemination and organization in mobile wireless environments. However, limited resources of wireless devices bring unique challenges to data access and information sharing. To address these challenges, we introduce the concept of content caching networks, in which the collected data will be stored by its contents in a distributed manner, while the data in the network is cached for a certain period of time before it is sent to a centralized storage space for backup. Furthermore, we propose a metadata-guided query evaluation approach to achieve query efficiency in content caching networks. By this approach, each cache node will maintain the metadata that summarizes the data content on itself. Queries will be evaluated first on the metadata before on the cached data. By ensuring that queries will only be evaluated on relevant nodes, the metadata-guided query evaluation approach can dramatically improve the performance of query evaluation. We design efficient algorithms to construct metadata for both numerical and categorical data types. Our theoretical and empirical results both show that our metadata-guided approach can accelerate query evaluation significantly, while achieving the memory requirements on wireless devices.