Scalable End-user Access to Big Data
May 16, 2013 2887 views 
Scalable end-user access to Big Data is critical for effective data analysis and value creation: Engineers in industry spend a significant amount of their time searching for data that they require for their core tasks. In the Oil&Gas industry, for instance, 30–70% of engineers’ time is spent looking for and assessing the quality of data. Optique is a large-scale European project focusing on the comprehensive and timely end-user access to very large data sets. In this presentation we describe how the concepts and technologies developed in Optique bring about a paradigm shift for data access by *providing a semantic end-to-end connection between users and data sources, *enabling users to rapidly formulate intuitive queries using familiar vocabularies and conceptualizations, *seamlessly accessing data spread across multiple distributed data sources, and thus *reducing the turnaround time for information requests from days to minutes. In the talk we will discuss the key concepts behind the Optique platform, including *the central role of ontologies and declarative mappings to capture user conceptualizations and to transform user queries into highly optimized queries over the data sources, *integration of distributed heterogeneous data sources, including streams, *the exploitation of massively parallel technologies and holistic optimizations to maximize performance, *tools to support query formulation and ontology and mapping management, and *semi-automatic bootstrapping of ontologies and mappings and query driven ontology construction to minimize implementation overhead. We will illustrate the value of the Optique platform through use cases from the case study partners in the energy domain: Siemens and Statoil.