Efficient storage and temporal query evaluation in hierarchical data archiving systems

Data archiving has been commonly used in many fields for data backup and analysis purposes. Although comprehensive application software, new computing and storage technologies, and the Internet have made it easier to create, collect and store all types of data, the meaningful storing, accessing, and managing of database archives in a cost-effective way remains extremely challenging. In this paper, we focus on hierarchical data archiving that has been popularly used in the scientific field and web data management. First, we propose a novel compaction scheme for archiving hierarchical data. By compacting both data and timestamps, our scheme substantially reduces not only the amount of needed storage, but also the incremental archiving time. Second, we design a temporal query language to support data retrieval from the compact data archives. Third, as compaction on data and timestamps may bring significant overhead to query evaluation, we investigate how to optimize such overhead by exploiting the characteristics of the queries and of the archived hierarchical data. Finally, we conduct an extensive experimentation to demonstrate the effectiveness and efficiency of both our efficient storage and query optimization techniques.