Sharing geoscience algorithms in a Web service-oriented environment (GRASS GIS example)

Abstract

Effective use of the large amounts of geospatial data available for geospatial research and applications is needed. In this paper, the emerging SOAP-based Web service technologies have been used to develop a large number of standard compliant, chainable geospatial Web services, using existing geospatial modules in software systems or specific geoscientific algorithms. A prototype for wrapping legacy software modules or geoscientific algorithms into loosely coupled Web services is proposed from an implementation viewpoint. Module development for Web services adheres to the Open GIS Consortium (OGC) geospatial implementation and the World Wide Web consortium (W3C) standards. The Web service interfaces are designed using Web Services Description Language (WSDL) documents. This paper presents how the granularity of an individual existing geospatial service module used by other geoscientific workflows is decided. A treatment of concurrence processes and clustered deployment of Web services is used to overcome multi-user access and network speed limit problems. This endeavor should allow extensive use of geoscientific algorithms and geospatial data.

Introduction

The rapid development of network technology has led to an explosion of data and information. For example, the National Aeronautics and Space Administration (NASA)’s Earth Observing System (EOS) satellites alone generate more than 3 terabytes (TB) of interdisciplinary earth system science data daily (NASA, 2007). Large volumes of geospatial data are archived at a variety of levels of processing. These Earth surface observation can be used for many purposes, among them are hydrological analysis, wildfire prediction, air pollution monitoring, water and land resource management and agricultural production estimation.

Each year a massive amount of funding is spent to collect, process, manage, and publish large amounts of geospatial data, but large amounts of the data have not been studied yet, for several reasons. Data processing is an additional burden and requires extensive processing time. The users are not familiar with the data format. The traditional research systems may involve high costs. One of the biggest problems is how to make the geospatial data readily available to scientists, policy makers, and the general public.

The accumulating data should be used effectively for geospatial research and applications. Geographic Information Systems (GIS) have come to be an indispensable tool for geospatial analysis and interpretation. They help the users to analyze, manage, and interpret spatial data. They facilitate geographic analysis and geographic modeling by scientists, policy makers, and the general public. A variety of proprietary and open source GIS products have been developed to meet practically all the basic needs of GIS users. Some modules to solve specific geoscientific problems are embedded into the GIS software. For most users, either the processing functions they want to use may be just a small part of a software package or some processing functions they need may not be provided in that software package. Moreover, many geoscientific algorithms or models use earth observation surface data to solve different kinds of earth sciences problems. Generally, these geoscientific algorithms or models created by different specialists are so complex that the users require considerable time to learn how to use them. Traditionally, these geospatial modules or models are developed and used on a local machine. This situation limits the use of geospatial algorithms or models, and potentially limits the use of geospatial data in modeling and analysis. The development of Web service technologies and new interoperability standards provides a solution to the above problems. The use of Web service technologies fulfills two crucial roles: the interoperability of Web services accelerates the sharing of spatial data and information by applications on different platforms, and the modularity of Web services promotes the sharing of specific geospatial functions by a wide range of users.

Web services are implemented by adopting a collection of communication standards for the global information society. The most critical technologies for implementation of Web services are eXtensible Markup Language (XML) (Bray et al., 2008), Simple Object Access Protocol (SOAP) (Mitra and Lafon, 2007), Web Services Description Language (WSDL) (Christensen et al., 2001), and universal description, discovery and integration (UDDI) (OASIS, 2004). At its most abstract level, a Web service consists of an XML document sent to and/or received from a remote software program (Newcomer, 2002). A service requester delivers a request message in the form of an XML document to a Web service and receives a response message in the form of an XML document from the Web service. XML hides the details of the underlying transport protocols and provides a platform-independent structured information format (Zhao et al., 2007). SOAP can encode messages to be exchanged between service requesters and providers. It provides a standard and simple way to structure XML messages. WSDL is an XML-based schema for describing the public interface (input/output parameters, and functionality) of a Web service in a standard and extensible manner. It is not intended for humans to read. It is generally used to support interoperable machine-to-machine interaction. Some information can be obtained from a WSDL document, including the data types to be exchanged, the operations offered, the protocol used, as well as the name and location of the Web service. UDDI is a protocol for publishing and discovering metadata about Web services. It allows the service provider to register a Web service in the Catalog Service for the Web (CSW) and facilitates the service requester's discovery and access to a Web service.

It is worth differentiating geospatial Web services as defined by the Open GIS Consortium (OGC) from Web services in the e-business world. OGC geospatial Web services have been developed in parallel with the evolution of Web services defined by the World Wide Web Consortium (W3C) and the organization for the advancement of structured information standards (OASIS). The geospatial Web services discussed in this paper are implemented using OGC and W3C standards. The reason to choose OGC standards is that OGC is the only international organization dedicated to develop geospatial implementation standards consistent with the international organization for standardization (ISO), the federal geographic data committee (FGDC), and the International Committee for information technology standards (INCITS), and other abstract or content standards. OGC standards have been fully tested in both NASA’s and other data providers’ environments. OGC specifications are widely used by geospatial communities for sharing data and resources and are becoming ISO standards (Deng and Di, 2009). When there is no relevant OGC standard, W3C Web services standards will be adopted.

In a broader sense, geospatial Web services embrace Web services that handle geospatial data and provide services in processing geospatial data (Zhao et al., 2007). OGC has published a series of geospatial Web services specifications for data sharing and processing. The Web Map Service (WMS), Web Feature Service (WFS), and Web Coverage Service (WCS) standards are widely used over the Web. Use of these services makes geospatial data more easily available. For clarity, this paper refers only to the customized geospatial Web services complete certain operations as “Web services”. Most geospatial Web services that have been implemented are classified as “processing services”, a taxonomic group following ISO/DIS 19119 (ISO/TC 211, 2005). Geoprocessing algorithms can be utilized effectively and shared using the Web service technologies.