Data Broker to Maximize Usage of Distributed Databases A case for agricultural decision support

1
Data Broker to Maximize Usage of Distributed
Databases-A case for agricultural decision
support-

• Seishi Ninomiya and Matthew Laurenson
• National Agriculture Research Center
• National Agriculture Research Organization
• ????????????????????????
• ????M ??????
• snino_at_affrc.go.jp

2
Features Agricultural Data/DB

• Small scale, various and distributed
• Heterogeneous similar databases
• Necessity of integration of various DBs to make
  decision
• Difficulty of centralization
• Several languages

Best Matches to Distributed DB System, GRID
3
Concept of GRID

• Maximize data and application usage without
  centralization
• Easy updates of data and applications
• Easy addition of new data and applications
• Flexible and dynamic integration of data and
  applications

4
Dynamical Integration of Separate DBs

• e.g. Pest Management System

Pest Management System
Pest Illustrated Reference
Pesticide Information
Pest Management Guide
5
Integrated Pest Management System
Pest Illustrated Reference
Pest Management Guide
Pesticide Information
Pest Forecasting
Pest Management System
.
6
But, DB heterogeneity causes

• e.g. Meteorological Databases

Heterogeneous Databases
Growth Predictions Model
Program for Data A
Weather DB A
Program for Data B
Weather DB B
Program for Data C
Program for Data A
Weather DB C
Program for Data B
Pesticide Prediction Model
Program for Data C
Very low efficiency of database use and
application development
7
Concept of Distributed DB
Soil Data
8
Data Broker brings

• e.g. Meteorological Databases

Heterogeneous Databases
MetBroker
Growth Prediction Model
Weather DB A
Disease Prediction Model
Weather DB B
Meta DB
Weather DB C
Farm Management Model
Weather DB D
New plug-in for new DB
Do not need to modify applications for new
weather DB
Very high efficiency of database use and
application development
9
Data Brokers

• Meteorological DB MetBroker (ready)
• Soil DB Broker (prototyped)
• Digital Elevation DB Broker (ready)
• Crop DB Broker (designed)
• Multilingual Resource Server (partially ready)
• Map DB Broker (ready by the other group)

10
MetBrokerWeatheer Databases currently accessed
11
Crop model applets with MetBroker


Disease prediction model
Pear growth model
Rice growth model
12
Weather DB Broker MetBroker
Implementations

• Uses RMI for communication with clients
• SOAP interface is also ready
• Uses JDBC, HTTP etc to access databases
• Demo applet and documentation at
  http//www.agmodel.net/MetBroker/MetBroker.html
• Source code is openly available underGNU Lesser
  General Public License

13
Spatial Query

• Request for a latitude longitude box
• Results from multiple weather databases for the
  requested area at once

Client
Broker
Temperature
Rainfall
Radiation
14
MetBrokerMapServer

• An application to display weather data on map

Wakayama Prefec. Data Amedas
Seoul University Data
http//www.mapblast.com
MetBroker
Weather Database
Map Database
15
Multilingual Support

• All MetBroker applications can be run in Korean,
  Thai, English, Japanese, Norwegian
• Adding a new language takes a few hours

16
Mobile Phone Interface

Mobile Phone with Web browser and e-Mail
17
Filed Monitoring Server with Wireless LAN
18
AG(agriculture)-Grid
Anywhere
User
?????
Enquiry
Answer
Info-Broker

• Multi-Lingual
• Ontology
• Translation system
• - Dictionary/Thesaurus

Diversity Language/Character/Culture
Sharing
Different Structure/Data set/DB system
19

• AFITA2002 will be held in Beijing from October 26
  to 28, 2002.
• Asian Federation for Information Technology in
  Agriculture

20
Thank you for your attention
http//www.agmodel.net http//agrinfo.narc.affrc.g
o.jp