Cloud- and Agent-Based Geospatial Service Chain: A Case Study of Submerged Crops Analysis During Flooding of the Yangtze River Basin

More intelligent construction of geospatial service chains and more efficient execution of such service chains remain major challenges in distributed geospatial analysis. This study addresses these challenges using a Cloud- and agent-based approach for automatic and intelligent construction of a geospatial service chain in the Cloud environment. A spatial agent infrastructure comprising fundamental services and an agent interface is designed, implemented, and deployed. Our approach involves a strategy for selecting and aggregating appropriate agents and Web-processing services (WPS) by evaluating their availability. This strategy ensures successful construction of a geospatial service chain in the Cloud environment, even when there is a lack of requisite geospatial services in the system. Moreover, the method can significantly increase the speed of a service chain in distributed environments and retains high stability when more requests are submitted over various network conditions. This is because the computing mobility and intelligence of the agent help to avoid transfer of large volumes of spatial data and keep the load balanced during construction and execution of the service chain. A prototype system for analysis of submerged crops during flooding of the Yangtze River basin demonstrates the advantages of our approach over existing methods.