1. 1 www.d4science.org Building scientific Virtual Research Environments in D4Science Paul Polydoras p.polydoras@di.uoa.gr University of Athens, Greece

2. 2 www.d4science.org Interoperability & VRE Definition Interoperability:"The capability to communicate, execute programs, or transfer data among various functional units in a manner that requires the user to have little or no knowledge of the unique characteristics of those units, ISO/IEC 2382-01, Information Technology Vocabulary, Fundamental Terms VRE: a digital environment that supports researchers across disciplines in the process of creation, validation and exploitation of data, information and knowledge as individuals and as groups, supporting collaboration and fulfilling their needs in ICT resources.

3. 3 www.d4science.org VRE : Interoperability elements on a domain-agnostic e- Infrastructure Resource Software Data Policies

4. 4 www.d4science.org Resources & Software Resources (Virtualisation & resource model (representation and use)) Resource Virtualization Common Resource Management Facilities Lifecycle Publication/Registration Monitoring Discovery Employing / Querying Software (empowering resources) Interface-based architecture Cooperate on interfaces, compete on implementations Loosely coupled components Message exchange and handling Management Compliance with Open Standards

5. 5 www.d4science.org Data & Policy interoperability Data interoperability Schema agnostic data handling Tolerant for structured, semi-structured and unstructured data (Meta) Data / Content interoperability Brokerage Data Transformation Policy interoperability Conceptualization Interoperating policy enforcement mechanisms

6. 6 www.d4science.org D4Science Reference Model Resource Model Information Model Policy Model

7. 7 www.d4science.org Resource Model

8. 8 www.d4science.org Information Model Reference Info-Object Property has references with Name has from to (0,n) (1,1) (1,n) Value OID Type Name Primary role Secondary role (0,n) content Position

9. 9 www.d4science.org VO Model

10. 10 www.d4science.org Focus on Data Interoperability in gCube Makes excessive use of XML data Provides dynamically constructed highly distributed data processing pipe-line Players are all service resources Provides the mechanisms for interoperable services exchanging data Data transfer not effectively captured by WS-* ResultSets Is data agnostic in the core: Metadata Management, Indexing, storage do not pose restrictions on payload. Exposes excessive configurability for handling domain-specific requirements Offers services for converting data and content among interoperating parties: The Metadata Broker The gCube Data Transformation Service

11. 11 www.d4science.org Focus on Data Interoperability in gCube : On the road Supports multiple protocols for importing / exporting data Soon OAI-PMH compliant WS-DAI compliance gCube concepts directly map to WS-DAI ones OAI-ORE compliance gCube Information model directly matches the ORE model Advanced data interoperability techniques, based on ontologies and inference

12. 12 www.d4science.org D4Science System Reference Architecture

13. 13 www.d4science.org Sum up D4Science/gCube provides machinery for: Operating an e-Infrastructure that supports Resource Sharing Design, Creation, Management for Virtual Research Environments Its targeted execution environment / scope dictates an inherent interoperable philosophy / architecture Every aspect (both model-based and systemic) is designed to be interoperable in principle

14. 14 www.d4science.org Thank you!