The Global Environment for Network Innovations (GENI) is a virtual laboratory at the frontiers of network science and engineering. Using a combination of widely-distributed resources connected with bandwidth-on-demand regional and global networks, GENI enables the worldwide research community—including academia, industry and the public—to test ideas and clean-slate designs to envision future internets at scale.
RENCI, in collaboration with Duke University, participates in the development of GENI architecture, maintains a portion of the GENI testbed, called ExoGENI and develops novel uses for the GENI infrastructure for data-intensive computational science domains.
The ExoGENI project, part of the National Science Foundation’s Global Environment for Network Innovation (GENI) initiative, is a nationwide test bed for advanced networking and networked cloud computing. ExoGENI links GENI to two advances in virtual infrastructure: open cloud computing (OpenStack) and dynamic circuit fabrics.
In effect, ExoGENI is a widely distributed networked infrastructure-as-a-service (NIaaS) platform geared towards experimentation and computational tasks. It brings together computation, storage and network capabilities on a global scale to address emerging complex problems related to network science, large-scale distributed computations, large data set mobility, and future network architectures. Using the ExoGENI test bed, networking researchers are able to allocate private networks spanning the globe, allocate computing clusters, storage, and other resources for use by scientists who collect and analyse data, and tie these resources to production networks, devices, or instruments.
ExoGENI server racks are deployed at 20 sites across the U.S., as well as in Europe, Australia and South America. ExoGENI is controlled by a deployment of ORCA (Open Resource Control Architecture), a network control framework originally developed at Duke, and further developed by RENCI for GENI needs and requirements.
- Ilya Baldin (Project Lead)
- Jeff Chase, Duke University (Co-PI)
- Mert Cevik
- Yufeng Xin
- Anirban Mandal
- Victor Orlikowski
- Paul Ruth
- Claris Castillo
GENI Science Shakedown works toward enabling GENI technologies in order to transform the way experimental researchers in domain sciences use networking and distributed systems. The GENI test beds (ExoGENI and InstaGENI) have matured to the point where their capabilities in providing domain science applications “at scale” can be evaluated. The results of this evaluation will inform the direction of future GENI development.
GENI Science Shakedown will determine whether the current GENI technologies can effectively support the computing needs of domain science and will evaluate the benefits and limitations of current GENI technologies. In addition, the project provides an opportunity to introduce domain scientists to the GENI view of networking and distributed systems in order to help disseminate GENI technologies to existing and future high-throughput and high-performance computing centers.
RENCI experts will apply the GENI experiment workflow to domain science applications and evaluate GENI’s ability to run domain science experiments, in terms of both performance and applicability. Applications to be studied include biology workflows and the ADCIRC storm surge modeling system.
- Paul Ruth (Project Lead)
- Anirban Mandal
- Jeffrey Tilson
- Brian Blanton