Virtual experiments in data intensive research

Organization and management of virtual experiments (VE) in data intensive research (DIR) has been widely studied in the several past years. The authors survey existing approaches to deal with VEs and hypotheses and analyze VE management in a real use-case from the astronomy domain. A review of existing systems that can act as platforms for conducting a VE has been carried out. Requirements for a system to organize VEs in data intensive domains have been gathered and overall structure and functionality for system running VEs are presented. The relationships between hypotheses and models in VE are discussed. The authors also illustrate how to model conceptually VEs, respective hypotheses, and models. Potential benefits and drawbacks of such approach are discussed, including maintenance of experiment consistency and reduction of potential number of experiments.