EUBRAZILOPENBIO - Infrastructure enhancement and optimisation final report (D4.3)

The EUBrazilOpenBio project aims to design a novel Data and Cloud Computing e-Infrastructure by integrating existing EU and Brazilian technologies and resources to support the processing demands of two biodiversity-related problems that are both data intensive and computationally intensive. This document is the second report on 'Infrastructure Enhancement and Optimization' and aims to describe the analysis and planning behind the integration of these technologies and resources to achieve the required performance objectives for the infrastructure. In particular, the report focuses on the activities and results obtained from applying and evaluating different cloud programming models, management strategies, and infrastructure technologies, in the context of the two Use Cases. The first report discussed the initial technological planning (described in in [1]) behind the definition of an appropriate configuration of both the technology and the infrastructure to serve the needs of the two Use Cases. The results of a detailed performance analysis of the original proposal have been used to help identify the set of enhancements and optimisations to improve the first prototype of the infrastructure that included data and computational technologies (described in [2][3]5]) as gCube, openModeller, EasyGrid AMS, COMPSs, Usto.re, VENUS-C and HTCondor. In this follow up report the advancement in the developments of the EUBrazilOpenBio infrastructure is described, detailing how the previously identified building components have been put together in order to implement the Use Cases. The results of the optimizations are discussed and a development plan for the production infrastructure is presented.