Enabling future internet research: the FEDERICA case

Abstract

The Internet, undoubtedly, is the most influential\ud technical invention of the 20th century\ud that affects and constantly changes all aspects\ud of our day-to-day lives nowadays. Although it\ud is hard to predict its long-term consequences,\ud the potential future of the Internet definitely\ud relies on future Internet research. Prior to\ud every development and deployment project, an\ud extensive and comprehensive research study\ud must be performed in order to design, model,\ud analyze, and evaluate all impacts of the new\ud initiative on the existing environment. Taking\ud the ever-growing size of the Internet and the\ud increasing complexity of novel Internet-based\ud applications and services into account, the\ud evaluation and validation of new ideas cannot\ud be effectively carried out over local test beds\ud and small experimental networks. The gap\ud which exists between the small-scale pilots in\ud academic and research test beds and the realsize\ud validations and actual deployments in production\ud networks can be bridged by using\ud virtual infrastructures. FEDERICA is one of\ud the facilities, based on virtualization capabilities\ud in both network and computing resources,\ud which creates custom-made virtual environments\ud and makes them available for Future\ud Internet Researchers. This article provides a\ud comprehensive overview of the state-of-the-art\ud research projects that have been using the virtual\ud infrastructure slices of FEDERICA in\ud order to validate their research concepts, even\ud when they are disruptive to the test bed’s infrastructure,\ud to obtain results in realistic network\ud environments.Postprint (published version

The FEDERICA infrastructure and experience

“…During the operational lifetime of the project, since January 2009, the consortium partners approached and consulted more than 50 potential user groups all over Europe that resulted in 20 slice requests for specific user experiments, with 5 slices active at the same time on aver-age. The article [5] provides a comprehensive overview of the research projects that have been using the virtual infrastructure slices of FEDERICA until 2011 to validate their research concepts, even being disruptive to their testbed slice, to obtain results in realistic network environments. The projects may be grouped according to their main objective; validation of virtual infrastructure features, evaluation of multilayer network architectures, and design of novel data and control plane protocols at various layers.…”

“…The slice contained more than one hundred resources in the case of the PHOSPHORUS project (see [5]), with an average of 20 virtual resources in a slice. FEDERICA has always been able to create the slices as requested.…”

“…The FEDERICA project (Federated E-Infrastructure Dedicated to European Researchers Innovating in Computing Network Architectures) [2][3][4][5][6] has been designed and operated to support research and experimentation on current and Future Internet. The project consortium comprised twenty stakeholders from Research Institutes, Universities, National Research and Education Networks (NREN) and industrial partners.…”

OpenFlow driven ethernet traffic analysis

“…To cope with this problem, projects like GENI [1] and FEDERICA [2] propose to slice network resources through network virtualization. In such architectures, it becomes possible to run experiments on a potentially large scale network within a given network slice without interfering with other slices.…”