![](bg6.png)
computers. The load modelled a different process.
The cardinality of data sets utilized during this
experiment equalled 20000. The results are
presented on figure 6.
In the future we consider extending the query
definition possibilities. One of the possible upgrades
is defining an area in which the query is executed.
This is a straightforward modification, which can
leverage the functionality and substantially reduce
the time required to obtain relevant results.
The experiment proved that executing queries in
a distributed environment is much more efficient
when the computers comprising the system are
simultaneously used for other tasks. It can utilize
computers shared by many queries less frequently
and move the main processing effort to computers
specific for each query. Such behaviour is
particularly desired when many queries are executed
in parallel.
REFERENCES
Yiu, M., L., Dai, X., Mamoulis, N., Vaitis, M. ‘Top-k
Spatial Preference Queries’, In proceedings of
International Conference on Data Engineering,
Istanbul, Turkey, 2007, pp. 1076 – 1085.
Guttman, A. ‘R-trees: A Dynamic Index Structure for
Spatial Searching’, In proceedings of ACM SIGMOD
International Conference on Management of Data,
Boston, USA, 1984, pp. 47 – 57.
0
50
100
150
200
250
no load 75% load
time [s]
AMD Duron 1,6GHz 640MB RAM Intel Pentium M 1,7GHz 1,5GB RAM
Intel Core 2 Duo 1,83GHz 1,5GB RAM distributed system
Gorawski, M., Dowlaszewicz, K. ‘Optimization of Top-k
Spatial Preference Queries’ Execution Process Based
on Similarity of Preferences’, In proceedings of
International Conference on Multimedia & Network
Information Systems, Wrocław, Poland, 2008.09, pp.
140 – 151.
Güntzer, U., Balke, W-t., Kießling W. ‘Towards Efficient
Multi-feature Queries in Heterogeneous
Environments’, In proceedings of International
Conference on Information Technology: Coding and
Computing, Las Vegas, USA, 2001, pp. 622 – 628.
Figure 6: Time of query execution using local L1P1S1 and
distributed DMDL algorithm while generating artificial
load on one of the computers.
Michel, S., Triantafillou, P., Weikum, G. ‘KLEE: A
Framework for Distributed Top-k Query Algorithms’,
In proceedings of International Conference on Very
Large Data Bases, Trondheim, Norway, 2005, pp. 637
– 648.
7 SUMMARY AND OUTLOOK
The paper presents a proposition of an algorithm for
execution of top-k spatial preference queries in a
distributed environment. The specifics of the query
and the target environment necessitated developing
adequate mechanisms that would ensure efficient
execution. Data processing was therefore split on all
nodes taking part in query execution and the number
of data transmitted through the network was
minimized. Apart from that further optimization
mechanisms were introduced. The paper discussed
both the specifics that motivated the development of
each technique and the technique itself. It also
presents an analysis of algorithm’s efficiency based
on conducted experiments. The results confirm that
the DMDL algorithm, which employs an efficient
auto adaptation method is capable of efficient
executing the queries in a heterogeneous distributed
environment. It can be utilized in a closed dedicated
system or in an open system comprising different
computers owned by different parties. The
experiments also proved that data can be distributed
on purpose in order to shorten the time of executing
complex top-k spatial preference by employing the
developed algorithm.
Bruno, N., Gravano, L., Marian, A. ‘Evaluating Top-k
Queries over Web-Accessible Databases’, In
proceedings of International Conference on Data
Engineering, San Jose, USA, 2002, pp. 369 – 380.
Gorawski, M., Dowlaszewicz, K. ‘: An Analysis and
Extension of Top-K Spatial Preference Queries
Optimization Methods’, In proceedings of
International Conference on Information Technology,
Gdańsk, Poland, 2008.05, pp. 227 – 230.
Hjaltson, G., R., Samet, H. ‘Distance Browsing in Spatial
Databases’, ACM Transactions on Database Systems,
vol. 24, issue 2, June 1999, pp. 265 – 318.
Braun, T., D., Siegel H., J., Beck, N., Bölöni, L.,
Maheswaran, M., Reuther, A. I., Robertson, J., P.,
Theys, M., D., Yao, B., Hensgen, D., Freund, R., F. ‘A
Comparison of Eleven Static Heuristics for Mapping a
Class of Independent Tasks onto Heterogeneous
Distributed Computing Systems’, Parallel and
Distributed Computing, vol. 61, 2001, pp. 810 – 837.
Munir, E., U., Li, J., Shi, S., Zou, Z., Yang, D. ‘MaxStd: A
Task Scheduling Heuristic for Heterogeneous
Computing Environment’, Information Technology
Journal, vol. 7, issue 4, 2008, pp. 679 – 683.
ICEIS 2009 - International Conference on Enterprise Information Systems
48