2011.12 WP11: Data e-Infrastructures Integration and Interoperability Facilities Development Monthly Activity By Task and Beneficiary
From IMarine Wiki
This activity report documents the activities performed in November and December 2011. It is integral to Activity Report of the period.
T11.1 Application Programming Interfaces Framework Definition
Partners: NKUA, CNR, CERN, E-IIS, FAO
During the first two periods, the task has been targeted around the assessment of current gCube APIs.
One activity of the task have been the validation of the perception built previously for the system API. This has been performed on new requirements for service integration in the area of Information retrieval. As a result the HTTP interface for searching, provided by the Application Service Layer, has been identified as a proper gateway through which applications targeting the provision of advanced data discovery in gCube, will be accessing the infrastructure data. The needs for evolution of this interface are investigated as part of T11.3.
Work has been done on an early definition of the architecture of Integration and Interoperability Layer, as an evolution of the current Application Service Layer, however this is more of an exercise rather than a concrete proposal for IIL.
Finally an early plan for the production of the D11.1 as well as the organization of the work in the WP has been prepared and remains to be presented to all partners for evaluation, fine-tuning and adoption early in the next period.
- No major issues have been faced at the stage of the definition of the principles and policies within the objectives of the task.
- A delay in the launch of coordination activities is observed due to the need of assessment of previous project's results. No corrective actions are foreseen currently as this is not expected to have an impact on mid-term deliveries of the WP.
T11.2 Data Management APIs
Partners: FAO, CNR, NKUA, CERN, E-IIS, FORTH, TerraDue
FAO has begun work on client APIs for the Data Management layer in the context of the forking of the Content Management subsystem into the Data Access subsystem. Two new components that replace legacy functionality in gCore have been defined:
common-clients, a gCore- and gCube-free library with basic abstractions for the definition of client APIs that connect to either individual service endpoints or else discover and connect to endpoints, caching them in both cases. Specific queries and endpoint reference types, are abstracted away in interfaces. Annotations are available and acted upon during the interaction strategies to make the semantics of faults.
common-gcore-clients, a gCore-bound library that specialises the abstractions in
common-clientsto gCore services, where queries are to the current IS, calls are scoped, and endpoint reference types are those current adopted.
The components are under currently under version control and build regularly along with their Software Archives in Etics.
common-gcore-clients, the new
tree-manager-library in the Data Access subsystem offers APIs for access to the new Tree Manager service.
- So far, FAO has given a personal interpretation to the overall methodology and design principles that may govern the WP. Given the nature of the WP suggests uniformity across solutions, we strongly recommend that, through WP leadership, such principles and methodology are governed by a shared plan.
- APIs for publication and discovery are foundational to any other client API. Waiting for output from T11.1, FAO has encapsulated these dependencies inside its own APIs implementations, which may or may not the agreed way forward. It is also likely that publication and discovery APIs may need to be directly exposed to clients, rather than serve solely for discovery strategies for APIs that target individual services.
- first two general-purpose components for client API development
- first API for data access through the new Tree Manager service
During this period NKUA work on data management has been in the direction of OGC APIs exposure and consumption by its underlying entities, mainly being the Geo-Index. Nevertheless work is premature and concrete decisions and products will emerge at later stages of the project.
none recorded yet
T11.3 Data Consumption APIs
Partners: CNR, NKUA, FORTH, TerraDue, FAO
In the area of Geo-spatial data access, NKUA has investigated thoroughly the area of OGC services. The investigation is looking for OCG standards applicable in the Search framework and has studied the cases of WCS and WCPS. This will be a long term results have been identified yet.
In an neighboring area, NKUA has investigated the query languages of multidimentional data management systems, targeting initially RasQL of the rasdaman system.
Regarding Search in general, all gCube search types (simple search, criteria based search, quick search, documents browsing) must be offered by the enabling elements of this layer. The gCube resources based on geographic extents must also be accessible through the OpenSearch and the adoption of the OpenSearch-Geo extensions to provide a standard query mechanism.
- Extension of gCube exposed OpenSearch interface to support OpenSearch-Geo extensions.