Differences From
Artifact [cab57e20a5]:
175 175 ** An example experiment
176 176 to show how the testbed works. We'll create two slivers which ping each
177 177 other. *##*
178 178
179 179 # Use summary diagram, maybe colorise labels.
180 180 1. The researcher first contacts the server and registers a slice description
181 181 which specifies a template for slivers (e.g. Debian Squeeze) and includes
182 - data and programs to setup slivers and run experiments. *##*
183 -2. This and all subsequent changes performed by the researcher are stored in
184 - the registry, which holds the config of all components in the testbed. *##*
185 -3. The researcher chooses two nodes and registers sliver descriptions for them
182 + data and programs to setup slivers and run experiments.
183 +
184 + The researcher chooses two nodes and registers sliver descriptions for them
186 185 in the previous slice. Each one includes a public interface to the CN.
187 - Then the researcher tells the server to instantiate the slice. *##*
188 -4. Each of the previous nodes gets a sliver description for it. If enough
186 +
187 + This and all subsequent changes performed by the researcher are stored in
188 + the registry, which holds the config of all components in the testbed.
189 +2. The researcher tells the server to instantiate the slice.
190 +
191 + Each of the previous nodes gets a sliver description for it. If enough
189 192 resources are available, a container is created by applying the sliver
190 - configuration over the selected template. *##*
191 -5. Once the researcher knows that slivers have been instantiated, the server
192 - can be commanded to activate the slice. *##*
193 -6. When nodes get instructions to activate slivers they start containers. *##*
194 -7. Containers execute the setup & run programs provided by the researcher. *##*
195 -8. Researchers interact straight with containers if needed (e.g. via SSH) and
196 - collect results from them. *##*
197 -9. When finished, the researcher tells the server to deactivate and
198 - deinstantiate the slice. *##*
199 -10. Nodes get the instructions and they stop and remove containers. *##*
193 + configuration over the selected template.
194 +3. Once the researcher knows that slivers have been instantiated, the server
195 + can be commanded to activate the slice.
196 +
197 + When nodes get instructions to activate slivers they start containers.
198 +
199 + Containers execute the setup & run programs provided by the researcher.
200 +4. Researchers interact straight with containers if needed (e.g. via SSH) and
201 + collect results from them.
202 +5. When finished, the researcher tells the server to deactivate the slice.
203 +6. And also to deinstantiate it.
204 +
205 + Nodes get instructions and they stop and remove containers, respectively.
206 +7. If the researcher wants to, the slice itself can be removed.
200 207
201 -This is a summary of all the previous steps. *##*
208 +This was a view of the testbed from a research perspective. From the
209 +community perspective, *##*
202 210
203 211 * Cooperation between community networks and Community-Lab
204 212 can take different forms. Given a typical CN like this, with most nodes
205 213 linked using cheap and ubiquitous WiFi technology: *##*
206 214
207 215 - CN members can provide an existing CD and let CONFINE connect a RD to it via
208 216 Ethernet. Experiments are restricted to the application layer unless the