A resilient methodology for accessing and exploiting data and scientific codes on distributed environments

The execution of data-based applications on distributed environments is prone to different failures in the different steps of the process. These steps range from the search on available references and data to the adaptation of the applications of interest to the different platforms where they are going to be efficiently run, which requires a deep understanding of their specific characteristics. In this work, a workflow to efficiently develop, maintain and execute highly portable distributed applications on dynamic environments, performing experiments based on Data Repositories, is presented. With this approach, the development, execution and maintenance of distributed applications is significantly simplified with respect to previous solutions, increasing their robustness and allowing running them on different computational platforms unattendedly. Data search and usage is also significantly simplified and can be automatically retrieved as input data into a code already integrated in the proposed workflow. Such a search is based on metadata standards and relies on Persistent Identifiers (PID) to assign specific repositories to the new produced output