3 | | This is an example topology of a campus OpenFlow network, that is configured to support the Pathlet experiment. '''This experiment will run through GEC12 (from now till November 30th).''' |
4 | | The configuration described here is in addition to the [wiki:OpenFlow/CampusTopology basic campus topology] setup for the Plastic slices project, '''but uses the same infrastructure, i.e. the host in the diagram is one of the existing myPlc hosts that is connected to an !OpenFlow switch in your campus.''' Currently only HP switches can support the Pathlet experiment, so if you only have an NEC or a quanta switch, please let us know. '''After the initial setup described in this page, only routine experimental support is expected from participating site contacts participating in this experiment (e.g. authorizing opt-in traffic). We will use this page to keep track of all relevant information for this experiment (configuration, IP addresses etc).''' |
| 3 | This is an example topology of a campus OpenFlow network, that is configured to support the Pathlet experiment. This experiment will run through GEC12 (from now till November 30th). |
| 4 | The configuration described here is in addition to the [wiki:OpenFlow/CampusTopology basic campus topology] setup for the Plastic slices project, but uses the same infrastructure, i.e. the host in the diagram is one of the existing myPlc hosts that is connected to an !OpenFlow switch in your campus. Currently only HP switches can support the Pathlet experiment, so if you only have an NEC or a quanta switch, please let us know. After the initial setup described in this page, only routine experimental support is expected from participating site contacts participating in this experiment (e.g. authorizing opt-in traffic). We will use this page to keep track of all relevant information for this experiment (configuration, IP addresses etc). |
16 | | In a nutshell they would like to run their pathlet routing algorithm, which is implemented as a Nox OpenFlow controller, to route their traffic between different GENI sites, where the links between sites are over the Internet, instead of the GENI backbone. '''They have already |
17 | | run experiments using their overlay deployment on general-purpose PlanetLab, but they can not compare these results with their in-network deployment because those nodes have no !OpenFlow support. The application they have been using for their experiments is a video streaming application, over UDP, with a maximum data rate of 200kbps.''' |
| 16 | In a nutshell they would like to run their pathlet routing algorithm, which is implemented as a Nox OpenFlow controller, to route their traffic between different GENI sites, where the links between sites are over the Internet, instead of the GENI backbone. They have already |
| 17 | run experiments using their overlay deployment on general-purpose PlanetLab, but they can not compare these results with their in-network deployment because those nodes have no !OpenFlow support. The application they have been using for their experiments is a video streaming application, over UDP, with a maximum data rate of 200kbps. |
99 | | '''Sites that might be involved with this experiment include : Clemson, Georgia Tech, GPO, Indiana, Rutgers, Stanford, Washington, Wisconsin.''' |
| 99 | Sites that might be involved with this experiment include : Clemson, Georgia Tech, GPO, Indiana, Rutgers, Stanford, Washington, Wisconsin. |