FUTEBOL UNIVBRIS User Manual Authors Carlos Colman Meixner – University of Bristol, UK Reza Nejabati – University of Bristol, UK Version 0.1 Abstract This document is a manual for users of the FUTEBOL University of Bristol (UNIVBRIS) testbed. It describes simple tutorial how to use the resources of UNIVBRIS testbed. Using those examples, the user will be able to build his/her own experiments. This project has received funding from the European Union's Horizon 2020 for research, technological development, and demonstration under grant agreement no. 688941 (FUTEBOL), as well from the Brazilian Ministry of Science, Technology and Innovation (MCTI) through RNP and CTIC.
18
Embed
FUTEBOL UNIVBRIS User Manual - ict-futebol.org.br · FUTEBOL UNIVBRIS User Manual Authors Carlos Colman Meixner – University of Bristol, UK Reza Nejabati – University of Bristol,
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
FUTEBOL UNIVBRIS User Manual
Authors Carlos Colman Meixner – University of Bristol, UK
Reza Nejabati – University of Bristol, UK
Version 0.1
Abstract This document is a manual for users of the FUTEBOL University of Bristol
(UNIVBRIS) testbed. It describes simple tutorial how to use the resources of
UNIVBRIS testbed. Using those examples, the user will be able to build
his/her own experiments.
This project has received funding from the European Union's
Horizon 2020 for research, technological development, and
demonstration under grant agreement no. 688941 (FUTEBOL),
as well from the Brazilian Ministry of Science, Technology and
The University of Bristol testbed is a virtual infrastructure (VI) formed by software-defined-network (SDN) controllers which are functional with a convergent orchestrator. The testbed is designed to perform experiments with Open Flow protocol between optical switches and ethernet packet switches. In this section we introduce; (i) the available components of the testbed; (ii) the physical topology and logical topology of the testbed; and (iii) the experiment environment from the side of the user.
2.1 – Available components
The testbed has hardware components and software components.
a) Hardware components
Each network component has a dispositive id, “dpid” and an OpenFlow ID.
In this sub-section we introduce the main functionalities from JfFed experimenter frequently
used in the experiments (Fig.4).
Figure 4: Logical topology.
a) New: creates a new experiment b) Open: opens an existing experiment. c) Open ESpect: opens the file with experiment specification. d) Save the experiment. e) Run, Update, Terminates or stops, and Recover an experiment.
3 – Experiment tutorial
In this section we introduce step-by-step a short tutorial for beginners and intermediate
experimenters.
3.1 – Defining an experiment
Step 1: Define the topology and resources for the experiment. We will connect three
OpenFlow switches to a pair of VMs. Figure 5 shows the example topology for this tutorial.
Switch NEC C
dpid: 05:00:00:00:00:00:00:03
CSEEDELPHICSEEBORSHA
NEC D
NEC A
NEC C
CSEEDELPHICSEEBORSHA
Switch NEC A
dpid: 05:00:00:00:00:00:00:01VLAN 57
Switch NEC D
dpid: 05:00:00:00:00:00:00:04
Packet Domain
Openflow ID: 360287970189639684
Openflow ID: 360287970189639683
Op enflow ID: 360287970189639681
GBE0/6
GBE0/22GBE0/16
GBE0/8
GBE0/5 GBE0/13
GBE0/15GBE0/16
Packet Domain
Figure 5: Physical and logical topology of the tutorial.
The header generated by default is essential for the experiment, please be sure that jFed
generate this header, if not please type it in the RSpec editor the example of Fig 9.
Figure 9: Header and RSpec initialization
Step 6: Define the sliver right after the beginning of the RSpec section. The structure follows the next example. The email of the administrator, description of the experiment, and then close sliver line (Fig. 10).
Figure 10: Sliver definition.
Step 7: Select the OpenFlow controller and group of resources available.
After the sliver line, define the OpenFlow controller using the ip-address 10.0.0.72 and TCP
port 6633. Some experiments can use primary and secondary controllers. In this example
we use only a primary controller and one OpenFlow “packet” group for ethernet protocol
layer 2 (Pag. 11). Please don’t forget to add the line closing the group.
Figure 11: Controller and group definition.
Step 8: Select the equipment to be use in the experiment.
In this step we add the three OpenFlow enabled switches following the physical topology
defined in step 1 (Fig.12). It is important to define each component and physical connectivity.
Step 10: Define the special network parameters be used by or between the switches. In this section we use the command “openflow:match”, to define the VLAN tag case between the Giga Ethernet switches or the wavelength to be used in the Optical Switches (See example in Fig. 15).
Figure 15: VLAN definition using the command match.
The set of commands “openflow:match” uses a “openflow:use-group” to associate the group
of equipment. In this example the group “packet” uses the VLAN 57 to assign flows. The
Step 11: Start running the Rspect script by clicking “run” icon. Before the “sliver” is submitted to the name and project need to be selected. The name is selected by the user, and the name in this case is “FUTEBOL”. The last step is to choose the duration of the experiment. The duration can be placed in days, hours, and minutes, as show in Fig.17.
Figure 17: Schedule
After the schedule is defined the Kick off screen (Fig.18), the command “start experiment” button will submit a sliver authorization request in the “OFAM”. Ones the administrator of the UNIVBRIS Testbed authorize the sliver the resources will be available for experimentation.
Figure 18: Kickoff screen.
This will be the last step after click on “Run” icon to submit the experiment. We call
OPENFLOW we will run in the project of FUTEBOL for 2 hours.The controller web interface
will show the following status after the switches are ready for experimentation (This diagram
will be soon available for experimenters through jFed).
3.4 – Defining a flow script.
Step 12: Define the flows between switches.
In this example we use a flow between VMs in cseedelphi to cseebosra using the three
The set of parameters of the command “match” and “instructions” add the rules for the flow; input port “in-port”, output ports, VLAN tag “vlan-match” and “apply-actions” into the traffic. The “action” defines how the traffic will flow between input and output ports, “output-node-connection” parameter is used. Finally, the “flow-name” defines the name of the applied rule. The rule for NEC A, is “in-port”: “6” and “output-node-connector”:”13”, means receive the flow traffic from port “6” and send it to the port “13”.