Database Reference
In-Depth Information
to communicate partial result or the join attribute values with its peers for every
join predicate present in the query. Thus each node has to do additional message
processing. This will effect the overall time taken by the query. But, with our
map structures we do not communicate for every join operation reducing the
communication delay. This less interaction between the database machines in the
cluster reduces the communication overhead and effect of poor load balancing.
Based on the data in Figure 5a to 5e, it is clear that, the query execution time
required by the state-of-the-art MySQL is more than our framework.
5Conluon
In this paper, we studied distributed aggregate query processing on cloud data
warehouses. We proposed storage structures PK-map and tuple-index-map, and
designed query-processing algorithm for processing these aggregate queries. We
analyzed the query processing, and advantages of using maps in eliminating sort
and group-by operations required by the aggregate operation. Our framework
not only reduces the communication cost but also makes minimum access to
relation tables depending on the type of aggregate operation. Results of our
extensive performance study demonstrate that the proposed approach improves
the performance of the ad-hoc aggregate query in Cloud Data Warehouses and
reduces communication overhead.
Acknowledgments. This work is supported in part by the NSF, under the
grants IIS-1149417, IIS-1239176, and IIS-1319600.
References
1. Kemper, A., Neumann, T.: Hyper: A hybrid OLTP and OLAP main memory
database system based on virtual memory snapshots. In: IEEE 27th International
Conference on Data Engineering (ICDE), Hannover, Germany, pp. 195-206 (2011)
2. Vlachou, A., Doulkeridis, C., Norvag, K., Kotidis, Y.: Peer-to-Peer Query Process-
ing over Multidimensional Data. Springer (2012)
3. Curino, C., Evan, P.C.J., Raluca, A.P., Malviya, N., Wu, E., Madden, S., Balakr-
ishnan, H., Zeldovich, N.: Relational Cloud: A Database Service for the cloud. In:
5th Biennial Conference on Innovative Data Systems Research (CIDR), California,
USA, pp. 235-240 (2011)
4. Simmen, D., Shekita, E., Malkemus, T.: Fundamental Techniques for Order
Optimization. In: ACM SIGMOD International Conference on Management of
Data, Montreal, Canada, vol. 25, pp. 57-67 (1996)
5. Kossmann, D.: The state of the art in distributed query processing. ACM Com-
puting Surveys (CSUR) 32(4), 422-469 (2000)
6. Xin, D., Han, J., Li, X., Benjamin, W.W.: Star-Cubing: Computing Iceberg Cubes
by Top-Down and Bottom-Up Integration. In: 29th International Conference on
Very Large Data Bases (VLDB), Berlin, Germany, vol. 29, pp. 476-487 (2003)
7. Boukhelef, D., Kitagawa, H.: E cient Management of Multidimensional Data in
Structured Peer-to-Peer Overlays. In: 35th International Conference on Very Large
Data Bases (VLDB), vol. 35. Lyon, France (2009)
Search WWH ::




Custom Search