Seamless Infrastructure Customization and Performance Optimization for Time-Critical Services in Data Infrastructures
Workshop: The Eighth International Workshop on Data-Intensive Computing in the Clouds
Authors: Spiros Koulouzis (University of Amsterdam), Zhiming Zhao (University of Amsterdam)
Abstract: The increasing volumes of data being produced, curated, and made available by research infrastructure (RI) initiatives in the environmental science domain require services to optimise the delivery and staging of data on behalf of investigators and other users of scientific data. Specialised data services for managing data lifecycle, for creating and delivering data products, and for customised data processing, play a crucial role in RIs to serve their user communities. We describe our experiences identifying the time-critical requirements of environmental scientists making use of ICT research support infrastructure. We present a microservice based infrastructure optimisation suite called the Dynamic Real-time Infrastructure Planner (DRIP). We provide a case study whereby DRIP is used to optimise runtime service quality for a data subscription service provided by the Euro-Argo RI using EGI FedCloud and EUDAT’s B2SAFE service.