This award funds risk reduction activities for a future, experimental network infrastructure that will allow researchers from diverse disciplines across computer and information science and engineering as well as from economics and the social sciences to escape today's Internet-circumscribed research environment. This infrastructure will allow researchers to program their own unique networks (i.e., multiple end-to-end "slices") that run on federated, heterogeneous backbones and edge networks to try out new protocols (that may or may not be TCP/IP-based), new architectures inspired by recent discoveries in economic game theory or artificial intelligence (that didn't exist when the Internet was first imagined), and/or new cross-layer research whereby functionalities that incorporate human values, such as information privacy or security, for example, become integral parts of the network architecture as opposed to an afterthought. The intention is for this infrastructure to be deeply instrumented so that researchers will be able to monitor their ongoing experiments, collect data on the use of novel protocols, designs and architectures, and analyze the emergent behaviors of traffic. Major outcomes from this project include 1) identification and reduction of technical risks, and 2) broadening and strengthening community participation in planning the experimental network infrastructure. Significant outreach activities will include the widest possible array of research and education communities in dialogue about the scope and scale of research on such an infrastructure, which promises to deepen their participation in the design process and helps to ensure that the future infrastructure meets their needs.
Overview GENI - the Global Environment for Network Innovations – is a suite of research infrastructure rapidly taking shape in prototype form across the United States. GENI aims to transform experimental research in networking and distributed systems, as well as emerging research into very large socio-technical systems, by providing a suite of infrastructure, as depicted in Figure 1, for "at scale" experiments in future internets. GENI supports two types of experiments. Controlled, repeatable experiments will greatly improve our scientific understanding of complex, large-scale networks. Alternately, "in the wild" trials of experimental services can engage large numbers of human participants within American universities and cities. GENI’s prototyping phase began when NSF awarded GENI Project Office (GPO) responsibility to BBN Technologies. Four years later, GENI has served over 600 experimenters. On behalf of these experimenters, GENI has satisfied more than 20,000 requests, drawing on the emerging nationwide GENI infrastructure deployed across US campuses, regional networks, and national research backbones (Internet2, NLR). GENI is a visionary undertaking with very high potential impact. If it succeeds, it will completely transform research in networking and distributed systems. Given the ever-growing importance of the Internet and global computing systems, GENI’s success could also create major opportunities for our economy and society. Years of community-driven effort have dramatically reduced the risk associated with bringing GENI to fruition. GENI Vision GENI’s vision is to become the world’s first laboratory environment for exploring future internets at scale, promoting innovations in network science, security, technologies, services and applications. GENI is now allowing academic and industrial researchers to perform a new class of experiments that tackle critically important issues in global communications networks: • Science issues – We cannot currently understand or predict the behavior of complex, large-scale networks. • Innovation issues – We currently face substantial barriers to innovation with novel architectures, services, and technologies. • Society issues – We increasingly rely on the Internet but are unsure that can trust its security, privacy, or resilience. Rather than build a separate, parallel set of infrastructure "as big as the Internet," which is clearly infeasible, we are GENI-enabling existing testbeds, campuses, regional and backbone networks, cloud computation services, and commercial equipment. GENI can then incorporate these networks and services by federation, rather than constructing and operating a separate set of infrastructure. Accomplishments We are extremely proud at how fast GENI is taking shape. After just four years of community prototyping effort, a working "meso-scale" GENI deployment is in active use, and a larger, "at scale" deployment is beginning. Figure 2 shows the current deployment, and proposed next steps to a larger, "at scale" nationwide GENI. This deployment includes GENI racks and GENI WiMAX. GENI racks bundle Software-Defined Networking (SDN) with virtualized computation and storage resources, bringing to the entire network the shared, deeply programmable environment that the cloud has brought to the datacenter. GENI-WiMAX virtualized cellular wireless communication technology provides an open, programmable, GENI-enabled "cellular-like" capability, expanding research avenues into the vital mobile side of next-generation networks. Real experimenters and educational users are the keys to the success of the GENI project. As GENI incorporates more capable and mature infrastructure, more experimenters have been able to use GENI for their research. One example is the web-based GENI Experimenter Portal, shown in Figure 3. Summary GENI is remarkable first for itself, as a fascinating new kind of system. Nothing like GENI existed three years ago. It is now up and running in prototype form across the US. This alone is a remarkable achievement. Far more important are the new classes of research that GENI has begun to enable. Hundreds of researchers are using GENI to investigate the future of global communications. As GENI gradually expands into its "at scale" form, these experiments will be able to explore an entirely new frontier in research, building up to at-scale experiments in future global communication systems that can engage tens or hundreds of thousands of human participants in university campuses across the United States. As GENI is being created, it is engaging an unusually wide range of participants from a cross-section of US institutions. This community is designing and developing GENI technology, deploying GENI on their campuses, and using GENI in their research. Several universities have begun applying GENI in their courses, with the potential to transform totally how experimental computer science is taught and learned. Perhaps most important to us, the GENI community, is that the newest generation of networking and distributed systems researchers have enthusiastically taken up research leadership roles in an ambitious project of national importance. We would like to congratulate the numerous academic and industrial groups – both Principal Investigators and the many graduate students – who are shaping GENI and turning it into a reality. Mark Berman, GENI Project Director Chip Elliott, GENI Futures Director
