From a classical web mapping publication to
a Inspire Service architecture
in the Ccloud
BRGM, Pierre LAGARDE
CNR, Claudio LUCCHESE
Currently, the majority of the environmental informationdissemination onthe web is done with the creation and the deployment of a specific map service on Internet (webmapping) in their the IT organization of the data provider. In the context of the European IST InGeoClouds project (www.ingeoclouds.eu), we experiment aim to change this paradigm and to develop a INSPIRE service-based INSPIRE approach for data publication in a scalable architecture without changing the users experience and data-publishers usual behaviours.
The InGeoClouds (CIP - ICT - PSP 2011 4.1 - Towards a cloud of public services) project aims at demonstrating the feasibility of employing a cloud-based infrastructure coupled with the necessary services to provide seamless access to geospatial public sector information, especially targeting the geological, geophysical and other geoscientific information. This kind of information exhibits interesting characteristics like the size of the available data, the existing metadata descriptions (mostly according to the European Directive INSPIRE) and the current availability of related services that can be moved to the cloud.
The poject targetsproject develops 3 objectivesrientations: a) a technical architecture dedicated to the geo-data publication in the public cloud but agnostic of the cloud provider, b) a definition and deployment of a scalable and “on-demand” application allowing to publish geo-information dedicated to the public sector, c) a functional approach to transform a map as a set of services in compliance with the requirements of INSPIRE regulations: metadata, view and download service.
The InGeoClouds (CIP - ICT - PSP 2011 4.1 - Towards a cloud of public services) project aims at demonstrating the feasibility of employing a cloud-based infrastructure coupled with the necessary services to provide seamless access to geospatial public sector information, especially targeting the geological, geophysical and other geoscientific information. This kind of information possesses interesting characteristics like the size of the available data, the existing metadata descriptions (mostly according to the European Directive INSPIRE) and the current availability of related services that can be moved to the cloud.
The presentation focuses on 3 challenges: elastic technical architecture, deployment of an “on-demand” geopublication service and functional approach.
The technical architecture is based on a set of scalable and independent components: elastic and scalable file system (glusterFSGlusterFS), replicated database in PostGres and high performance data publication WMS/WFS server (MapServer). The project defines and implements a specific REST API to manage each component, in particular to organize the elasticity of the architecture (autoscaling) and to simplify the account/users management by data-provider’s applications. The high-level API exposes REST services to interact easily with the geospatial components and manage the publication of OGC services.
This architecture is deployed in the public Amazon Web Service cloud. The project analyses the cost of this architecture in comparison with a classical “homein-premises” IT architecture and the benefits and weakness of an “on-demand” architecture for a SDI infrastructure. Preliminary results confirm that “moving to the cloud” can bring some interesting cost reduction. In addition, the deployed architecture can exploit the elasticity and scalability features of a cloud environment.
The second step achievement is to the deployment of facilities supportingapplications oriented to the process of data import and data publication approach (Geo-publication use case in the InGeoClouds project). The application rationalestrategy is to provide a “data-provider” applicationworkspace, to guarantee the “isolation” of each data pushed by a data-provider in his workspace within the cloud infrastructure, to manage the security of the system and to mask all technical challenges to respect the performance and availability of the OGC services required by the INSPIRE directive. The project will analyse the elasticity of the architecture to expose WMS and WFS services with stress testbeds of the application.
Instead of proposing a simple interface to transform geo-dataset as OGC services and to add this service to a common geo-portal (webmapping), the application defines a map-oriented design for a data-provider. Based on the content of the OGC Web Map Context standards (a ordered list of layers) with extensions to manage map options, like Man Machine Interactions, functionalities options, design interfaces, etc.…, the data-publisher focuses on the creation of an specific map with his data. But, in a same time, the application allows to “transform” the (web-) map as a set of INSPIRE services for visualize and download datasets. Data-providers, like local authorities, national organisations, surveys, associations, etc.,… are should requested to use a software/service (Software as a Service) to communicate publish their data on the web and to respect the INSPIRE requirements. The system we are presenting here goes in the direction of supporting this process.
In summaryconclusion, the InGeoClouds project proposes an alternative solution to create INSPIRE services without changing the principles of the directive, in particular data-provider responsibilities for publishing, sharing and re-using geospatial datasets as services in a distributed architecture (SDI). The definition of a geospatial extensible architecture, the use of the elasticity of the cloud infrastructure and the deployment of a data-provider software for publishing maps on the Web define an easily solution for public sector to be compliance with the INSPIRE regulations without investment in a specific in-house infrastructure.