Differences From
Artifact [795c276e2e]:
156 156
157 157 Finally, the server and nodes publish management information through an API
158 158 that can be used to study the testbed itself or to implement external services
159 159 (like node monitoring and selection).
160 160
161 161 ** An example experiment
162 162 # Event diagram, hover over components explained.
163 -To show how the testbed works: two slivers, one of them (source sliver) pings
164 -the other one (target sliver).
163 +To show how the testbed works: two slivers, one of them pings the other one.
164 +Let's call them the source and target sliver, respectively.
165 165
166 166 1. The researcher first contacts the server and creates a slice description
167 - which specifies a template for slivers (e.g. Debian Squeeze i386).
168 - Experiment data is attached including a program to setup the experiment and
169 - another one to run it.
170 -2. The server updates the registry which holds all definitions of testbed,
171 - nodes, users, slices, slivers, etc.
167 + which specifies a template for slivers (e.g. Debian Squeeze i386). The
168 + researcher attaches experiment data including a program to setup slivers
169 + for the experiments and another one to run them.
170 +2. This and all subsequent changes initiated by the researcher are stored in
171 + the registry, which holds the config of all components in the testbed.
172 172 3. The researcher chooses a couple of nodes and creates sliver descriptions
173 173 for them belonging to the previous slice. Both sliver descriptions include
174 - a public interface to the CN and user-defined properties for telling apart
175 - the source sliver from the target one. Sliver descriptions go to the
176 - registry.
174 + a public interface to the CN and user-defined properties to mark slivers as
175 + either source or target.
177 176 4. Each of the previous nodes gets a sliver description for it. If enough
178 177 resources are available, a container is created by applying the desired
179 178 configuration over the selected template.
180 179 5. Once the researcher knows that slivers have been instantiated, the server
181 - can be commanded to activate the slice. The server updates the registry.
180 + can be commanded to activate the slice.
182 181 6. When nodes get instructions to activate slivers they start the containers.
183 -7. Containers run the experiment setup program and the run program. The
184 - programs query sliver properties to decide their behaviour.
182 +7. Containers execute the experiment's setup and run programs. The programs
183 + query sliver properties to decide whether to act as source or target.
185 184 8. Researchers interact straight with containers if needed (e.g. via SSH) and
186 185 collect results from them.
187 186 9. When finished, the researcher tells the server to deactivate and
188 187 deinstantiate the slice.
189 188 10. Nodes get the instructions and they stop and remove containers.
190 189
191 190 * Cooperation between community networks and Community-Lab