International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B2. Istanbul 2004
1) Extract keywords from Q. By keywords matching in
information retrieval, we can carry out the GIS service
search process.
2) First, search the local services indexes at the local peer. If
there are services matching the query, then go to 3);
otherwise, go to 4).
3) The user see whether there are services (s)he wants by
checking services' descriptions that is returned. If there is
at least one service (s)he wants, then the process of GIS
service discovery is over; otherwise, go to 5)
4) Select randomly an initial link of the local peer. Then
clone and dispatch a worker agent with the GIS service
query to the peer at the other end of the selected initial
link. At that remote peer, doing the searching as at the
local peer.
5) Cloning a working agent and dispatching it with the
service query to the local peer's neighbour. At the
neighbour peer, doing the searching as at the local peer.
6) At the remote peer, once there are services matching the
query, then returning the matching services’ descriptions
to the user who decides whether the returned results
contain the target service. If the target service is found,
then the search task is over, and the working agent would
return the source peer or be destroyed at the remote peer.
If no target service is found, the working agent has to
continue the search target till the target service is found or
the working agent’s TTL is 0.
Note in the service discovery process above, when the working
agent gets to a peer along a neighbor link, its TTL will not
decrease; Only walking along initial link, its TTL will decrease.
4. A PROTOTYPE
In order to assess the feasibility of the architecture, a simplified
prototype is developed and some GIS Web services have been
implemented. We use the Aglet Software Development Kit 2.0,
J2SDK 1.4.2, BestPeer and Geotools 0.8.0 for implementing the
prototype. Hardware includes. five PCs, in which two PCs are
used for LIGLO servers and the other three PCs are used as BP-
GServices node. Geographic information of the states, cities,
rivers, roads, and lakes of the Canada from ARCView GIS 3.2
are split into three parts to be stored in the three the BP-
GServices nodes respectively and some basic GIS services are
provided. Figure 4 shows the GIS Web services of the prototype
and the query result for rivers, provinces and lakes.
= [Ox
| Roads_rt.shp =
i”) Roads.shp
AF| Rivers.shp
NÉ.
A| Province.shp
[—] 9045 - 0.46
[EZ] 0.15 5.157
EN 5.157 - 19.16-
13.184 - 29.3
HR 29.500 - 59.36
of] Lakes.shp
=m
| Drainageshp
A
/
| Chies.shp
=
Figure 4. A query result of the prototype
824
S. CONCLUSIONS
In this paper, we explore the techniques of establishing GIS
Web services systems in P2P environment. And a P2P based
GIS Web services framework is proposed. By combining Web
Services technology and P2P technology into GIS, we add more
flexibility and autonomy to GIS Web services systems, and
alleviate to some degree the inherent limitations of the
centralized systems. As an ongoing project, implementation of
BP-GServices is still underway. After the BP-GServices
prototype is finished, we're ready to integrate existing
geographic data and information into the system. We also plan
to extend BP-GServices to embrace semantic GIS Web services
in the near future.
REFERENCES
Ooi, B.C., Tan, K-L., Lu, H., et al,. 2002. P2P: harnessing and
riding on peers. In: Proceedings of NDBC 2002.China.
Gottschalk, K,. Graham, S, Kreger, LH. et: al, 2002.
Introduction to Web services architecture. /BM Systems Journal,
41(2), pp.170-177.
UDDI: http://www.uddi.org/ (accessed 15 Oct.2003)
Ne WS, Ooi.B.C, Tan, K-L., 2002. Bestpeer: A self-
configurable peer-to-peer system. In: Proceedings of ICDE2002,
San Jose, CA. pp.272.
BestPeer: http://xenal.ddns.comp.nus.edu.sg/p2p/. (accessed 10
Dec.2003)
Ng, W., Ooi, B.C. Tan, K-L. et al., 2003. PeerDB: a P2P-based
system for distributed data sharing. In: Proceedings of
ICDE2003. pp.633-644.
Ling, B., Lu, Z, Ng, W.S., et al, 2002. A content-based
resource location mechanism in PeerIS. In: Proceedings of
WISE2002. pp.279-290.
Wang, X., Ng, W.S., Ooi, B.C., et al., 2002. Buddyweb: a p2p-
based collaborative web caching system. In: Proceedings of the
2002 International Workshop on peer-to-peer computing.
pp.247-251
R. Baesa-Yates and B. Ribeiro-Neto.,1999. Modern Information
Retrieval. Addison Wesley, Reading, Mass.
C. Ng and K. Sia.,2002. Peer clustering and firework query
model. In: Poster Proceedings of 11th World Wide Web
Conference (WWW2002), Honolulu, Hawaii.
Cooper, B., Garcia-Molina, H., 2003. SIL: Modeling and
Measuring Scalable Peer-to-Peer Search Networks. In:
Proceedings of DBISP2P. pp.2-16.
Ganesan, P., Sun, Q., Garcia-Molina, H., 2003. YAPPERS: A
Peer-to-Peer Lookup Service Over Arbitrary Topology. In
Proceedings of INFOCOM 2003, San Franciso, CA, USA.
Graupmann, J., Biwer, M., Zimmer, P., 2003. Towards
Federated Search Based on Web Services. In: Proceedings of
10" Conference on Database Systems for Business, Technology
and Web (BTW), Leipzig, Germany. pp.384-393.
Intern
Zhou,
Servic
Sanya
Gnute
GML.
ro (ace
Aglets
J2sdk |
Geoto
The w
grant |
Progra
Resea
LIESN
