Frequently Asked Questions

• What is GENI?

  GENI, which stands for Global Environment for Network Innovations, will be a shared, global facility designed to catalyze research on network architectures, services and applications.

• How is GENI different from a traditional network testbed?

  Unlike traditional network testbeds that are designed to validate a specific proposal, GENI will be a general-purpose facility that places essentially no limits on the network architectures, services, and applications that can be evaluated. Unlike traditional network testbeds that limit researchers to either incremental changes or synthetic workloads, GENI will be designed to allow both clean-slate designs and experimentation with real users under real-world conditions. Unlike traditional testbeds that provide no credible deployment path to the commercial world, GENI will represent a model in which incremental adoption of new services drives wide-spread deployment.

• What are the main technical ideas exploited by GENI?

  At the physical level, GENI will consist of a collection of physical networking components, including links, forwarders, storage and processor clusters, and wireless subnets. These resources are collectively called the GENI substrate. On top of this substrate, the GENI management framework will overlay network experiments on the substrate, where each experiment is said to run in a slice of the substrate.

  Four key ideas make this possible. First, the substrate components will be programmable. This will make it possible to embed any network experiment in GENI, including clean-slate designs that are radically different from today's Internet architecture and protocols. Second, the substrate will be virtualizable. This will make it possible to embed multiple slices in the substrate at the same time. This is important because it will allow experimental services and architectures to run continuously, rather only in a reserved time slot. Third, GENI will include mechanisms that allow end-users to seamlessly opt-in to experimental services. This will make it possible to attract real users, which is critical to evaluating new services under real-world conditions. It also enables incremental adoption, which is key to encouraging wide-spread deployment. Finally, GENI will be modular, with a well-defined architecture and set of interfaces. This will make it possible to extend GENI with new networking technologies as they become available. That is, GENI will not be a static artifact, but rather a dynamic infrastructure that is continually renewed.

• Who is defining GENI?

  The stake-holders are still being identified. What we can say is that we expect to use a competitive process that draws upon strong university-industry-government partnerships.

• Who will build GENI?

  The expectation is that the research community will be largely responsible for building GENI, with some number of development teams spread across the country's Universities and research labs involved in its construction. These teams will design and implement the various hardware and software components that will collectively form GENI, leveraging commercially available technology when appropriate. Additional teams will be responsible for assembling the components into a coherent working system, and managing the system on behalf of researchers.

• Is GENI's definition set in stone?

  No. The proposed facility described in various pages of this web site represents a synthesis of the NSF workshops that have taken place over the past year. Planning processes requires that GENI be defined with a certain level of specificity, but everyone understands that the underlying technology is likely to change rapidly, and that the requirements the community places on GENI will continue to mature. Therefore, the overview of GENI should be viewed as a snapshot as of January 2006. This site will post future snapshots as they come into focus.

• How can I participate?

  Planning for GENI is still in the very early stages. The definition of the GENI facility will continue to be refined over the next several years, including after we begin to build it. Current snapshots of GENI's design will be posted on this web site, and feedback solicited through a discussion mailing list and a series of open town hall meetings. You are also encouraged to communicate directly with colleagues on various working groups.

• How will GENI be paid for and what impact will this have other NSF research programs?

  NSF is purusing new money (not currently in the CISE budget) for GENI. As the offical NSF web site for GENI suggests, funding for research using GENI will come from standard CISE programs, such as NeTS, Cyber Trust, CSR, and other programs, using standard procedures for determining which proposals merit funding.

• What research is GENI expected to enable?

  GENI is expected to enable research across several CISE sub-areas, including networking, distributed systems, wireless and sensor nets, and optical networking. This is consistent with the rationale for GENI, which considers a wide-range of challenges and opportunities facing the Internet.

  Note that the rationale for GENI is much broader than simply investigating alternatives to the Internet's core protocols (e.g., TCP/IP), and there is certainly no pre-conceived notion that the outcome will be limited to innovations at any one level of the protocol stack. GENI is equally likely to result in alternative protocols and architectures running inside the network, as in new applications and services running as overlays on top of today's network.

  Understanding how different types of experiments will influence (and stress) GENI's design is a subject of ongoing discussions. For a recent synoposis of the design principles guiding GENI's design, the interested reader is referred to a recent design document.

• I'm not a networking researcher. What will GENI do for me?

  Researchers not working on design issues directly supported by GENI will still find many related research questions inspired by GENI. These will include the theoretical underpinnings of complex network systems, analysis and modeling of data collected on GENI, and broader inter-disciplinary issues related to law, policy, and societal impact.

• Is the expectation that research done on GENI will result in a new version of IP?

  Perhaps, but not necessarily. We expect innovations at many levels of the protocol stack, not just the network level. It might be that GENI leads to the discovery of new overlay services or network applications that run on today's network, but will have a transformative effect on the Internet over time. This is analogous to the way the Internet had a transformative effect on the telephony system over the last 30 years. In general, there are many possible success scenarios worth considering.

• The deployed Internet represents an enormous investment. How is a clean slate design possibly going to replace it?

  GENI is designed to enable clean slate thinking that is not constrained by today's Internet, but this does not imply that the outcome will necessarily be an entirely new Internet. In other words, "clean slate" is a process, not a result. It is likely that different researchers will chose to leverage different aspects of today's Internet, while exploring alternatives to other elements.

  On the other hand, GENI is designed with deployment in mind. By lowering the barrier-to-entry for users that want to use new services and applications deployed on GENI, we hope to foster incremental adoption, which has the potential to lead to wide-spread deployment. Exactly how research enabled by GENI will ultimately influence the Internet is a subject of much discussion in the research community.

• GENI is suppose to be a global facility. How will other countries be involved in GENI?

  GENI's design will likely have explicit support for federation. This will allow similar facilities built by other countries to connect to GENI, thereby creating a network substrate with global reach. In such a design, each country or organization that contributes resources to GENI would retain autonomy over how those resources are used, but we are pursuing a design that will allow researchers to share this global infrastructure. As a starting point, GENI will integrate the resources of PlanetLab, which spans 300 sites across 30 countries.

• How might the larger scientific community take advantage of GENI?

  Again, because of its support for federation, we expect other research communities (and the government agencies that support them) will be able to connect their own resources (e.g., lambdas, compute clusters) to GENI, thereby gaining access to the new network services and applications deployed on GENI. These communities will be able to specify that the resources they contribute will primarily benefit their own user communities, with the researchers working on GENI benefiting from gaining access to real users.

• How will industry be involved in GENI?

  The networking community has a long history of collaboration with the computer and communications industry. We expect that to continue with GENI. Industry is likely to provide technologies that can be incorporated into the GENI substrate, but more importantly, we expect industrial partners to "plug" emerging technologies into GENI in a way that allows researchers to incorporate them in their experiments.

• What is the relationship between GENI and NSF's Cyberinfrastructure?

  GENI, and the research it supports, are complementary to NSF's broader CyberInfrastructure activities.

  GENI is a purpose-designed research infrastructure for a particular community, just like a telescope is for astronomers or a large sensor array in the ocean is for oceanographers. In this case, the research community to be served is the community of networking, distributed systems, and communications technology researchers.

  In contrast, the objective of the cyberinfrastructure program is to create broad-use production tools to be shared across the breadth of scientific research and education: tools such as visualization, data sharing, access to High Performance Computing capabilities, and similar facilities.

  The two programs are complementary because there is every reason to expect that research fostered by the GENI effort will directly impact and benefit NSF's CyberInfrastructure activities. Further, in cases where the CyberInfrastructure programs finds it appropriate to fund research in networking and related areas, GENI will provide a powerful and effective platform for supporting that research.

• What is the relationship between GENI and IPv6?

  First, it is important to remember that discussions about IPv6 began 15 years ago, with wide-spread deployment still just around the corner. With GENI, we may be at the early stages of a network architecture that will take root 10-15 years from now. Second, a new version of IP is not necessarily the only outcome of the research that takes place on top of GENI. It might be that GENI leads to the discovery of new overlay services or network applications that run on today's network, but will have a transformative effect on the Internet over time.

• What is the relationship between GENI and PlanetLab?

  PlanetLab serves as a prototype of GENI, and helps make the case that such a facility is feasible. However, GENI goes far beyond PlanetLab in several dimensions. Most importantly, instead of being limited to a set of commodity PCs running as an overlay on top of today's Internet, GENI will also support a richer set of node technologies and expose low-level network (link) behavior, thereby enabling experiments "inside" the network. It is also important that GENI provide a richer set of support services, lowering the barrier for researchers to be able to take advantage of GENI.

• What is FIND and how does it relate to GENI?

  FIND, which stands for Future InterNet Design, is an architecturally focused research effort that would benefit from the GENI facility. FIND defines a broad agenda for the NeTS community, and is likely to be a significant component of the GENI research program, but it does not necessarily capture all the research that will be enabled by GENI. FIND has a strong focus on defining a comprehensive architecture for the Future Internet, with research teams encouraged to reach consensus on broad architectural themes.