ORCHESTRATING GRID COMPUTING ENABLED WEB PROCESSING SERVICES

Existing Spatial Data Infrastructures (SDI) are mainly focused on data retrieval, data processing and data visualization. An open standard based – for example an Open Geospatial Consortium (OGC) based - SDI mostly supports the retrieval and visualization of data through web services. But the data processing is normally performed by humans with more or less proprietary and monolithic Geo Information Systems (GIS). With growing network capacity and processing power, some efforts were made to integrate stand-alone geoprocessing applications and their expert functionality into a web service environment and therefore enable web services to execute geoprocessing tasks. The OGC Web Processing Service (WPS), which became an official standard in late 2007, is a major attempt to address this issue in a standardized way. The WPS specification defines a standardized interface to publish and perform geospatial processes over the web. Such a process can range from a simple geometric calculation (for example a simple intersect operation) to a complex simulation process (for example a global climate change model).

To speed up the processing of large amounts of data and perform complex calculations (for example when doing a weather forecast simulation), the use of Grid Computing or related methods and technologies are a good choice for achieving high calculation performance, for improving service availability and for guaranteeing different Quality of Service (QoS).

Even though Grid Computing or Distributed Computing are no new approaches, only little research had been done on combining OGC Web Services (OWS) standards and implementations with such a manner.

However, the intrinsic complexity of geodata often requires the use of several processing steps to address a given problem. Grid Computing can be applied on one dimension to improve the performance of each step, but to automate the whole business process, orchestrated geoprocessing workflows have to be built. This would create a second dimension and enable the creation of high-speed and fully automated value-added geoprocessing workflows.

This paper will give an introduction to the new OGC WPS standards and will present an approach on how to combine this specification with Grid Computing. This will be combined with the introduction to an approach on how to orchestrate these gridified WPS in workflows.

Finally, the presented approaches will be validated by means of a real world scenario. A geoprocessing workflow will be modeled to solve the given problem in the field of fire protection in southern Spain. The Business Process Execution Language (BPEL) will be used to design this workflow, which will contain some gridified WPS Web Services. Modeling, execution and result visualization will all be performed in a single integrated environment and will prove the introduced approaches.

A list of the acronyms

SDI - Spatial Data Infrastructures

OGC - Open Geospatial Consortium

GIS - Geo Information Systems

WPS - Web Processing Service

QoS - Quality of Service

OWS - OGC Web Services

BPEL - Business Process Execution Language