Academic
Publications
Workload-Aware Indexing of Continuously Moving Objects

Workload-Aware Indexing of Continuously Moving Objects,Proceedings of The Vldb Endowment,Kostas Tzoumas,Man Lung Yiu,Christian S. Jensen

Workload-Aware Indexing of Continuously Moving Objects   (Citations: 5)
BibTex | RIS | RefWorks Download
The increased deployment of sensors and data communication net- works yields data management workloads with update loads that are intense, skewed, and highly bursty. Query loads resulting from location-based services are expected to exhibit similar character- istics. In such environments, index structures can easily become performance bottlenecks. We address the need for indexing that is adaptive to the workload characteristics, called workload-aware, in order to cover the space in between maintaining an accurate index, and having no index at all. Our proposal, QU-Trade, extends R- tree type indexing and achieves workload-awareness by controlling the underlying index's filtering quality. QU-Trade safely drops in- dex updates, increasing the overlap in the index when the workload is update-intensive, and it restores the filtering capabilities of the index when the workload becomes query-intensive. This is done in a non-uniform way in space so that the quality of the index re- mains high in frequently queried regions, while it deteriorates in frequently updated regions. The adaptation occurs online, without the need for a learning phase. We apply QU-Trade to the R-tree and the TPR-tree, and we offer analytical and empirical studies. In the presence of substantial workload skew, QU-Trade can achieve index update costs close to zero and can also achieve virtually the same query cost as the underlying index.
Journal: Proceedings of The Vldb Endowment - PVLDB , vol. 2, no. 1, pp. 1186-1197, 2009
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...Many recent works such as [4], [9], [16], [29], [34] focus on other specific problems and provide corresponding solutions...

    Yanbo Wuet al. P2P Object Tracking in the Internet of Things

    • ...Many recent works such as [3], [14] focus on other specific problems and provide corresponding solutions...

    Yanbo Wuet al. Tracing Moving Objects in Internet-Based RFID Networks

    • ...With a few notable exceptions [6, 19], past research on update performance in spatio-temporal indexing assumes that the data is disk resident [3, 11, 16, 18]...
    • ...However, new applications such as location-based services exhibit workloads that not only contain queries, but also contain frequent updates, studies now also consider update-intensive workloads [3, 11, 16, 18]...
    • ...We study performance on three modern platforms: a dual quad-core AMD Opteron 2350 (Barcelona) with 8 hardware threads, a dual quad-core Intel Xeon X5550 (Nehalem) with 16 hardware threads, and an 8-core Sun Niagara 2 (T2) 2 Empirical evidence suggests this is a strong assumption [18]...

    Darius Šidlauskaset al. Thread-Level Parallel Indexing of Update Intensive Moving-Object Workl...

    • ...QU-Trade adapts an R-Tree type moving object index to trade update for query performance in specific index regions according to the workload [55]...

    Jens Dittrichet al. MOVIES: indexing moving objects by shooting index images

Sort by: