43 | | Experimenters request resources from aggregates using a standard API called the [wiki:GeniApi GENI Aggregate Manager API] or GENI AM API. The AM API allow experimenters to, among other things, list the resources available at an aggregate, request specific resources from the aggregate be allocated to their slices, find the status of resources from the aggregate that are allocated to their slice and delete resources from their slice. The AM API uses [wiki:GeniRspec GENI Rspecs] (resource specifications) to describe resources. Most experimenters will not need to learn details of the AM API or read/write Rspec files; GENI experimenter tools hide much of this complexity. |
| 43 | Experimenters request resources from aggregates using a standard API called the [wiki:GeniApi GENI Aggregate Manager API] or GENI AM API. The AM API allow experimenters to, among other things, list the resources available at an aggregate, request specific resources from the aggregate be allocated to their slices, find the status of resources from the aggregate that are allocated to their slice and delete resources from their slice. The AM API uses [wiki:GeniRspec GENI Rspecs] (resource specifications) to describe resources. Experimenters send to aggregates a ''request'' Rspec that describes the resources they want and get back from the aggregates a ''manifest'' Rspec that describes the resources they got. The manifest includes information such as the names and IP addresses of the compute resources (e.g. virtual machines), user accounts created on the compute resources, etc. Most experimenters will not need to learn details of the AM API or read/write Rspec files; GENI experimenter tools hide much of this complexity. |
52 | | The following is the workflow for a typical GENI experiment. The objective here is to show how the concepts described above tie together; this is not intended to be a complete description of the GENI experimenter workflow. |
| 52 | The following is the workflow for a typical GENI experiment. The objective here is to show how the concepts described above tie together; this is not intended to be a complete description of the GENI experimenter workflow. The experimenter: |
| 53 | 1. Requests an account at the GENI Portal. |
| 54 | 2. Joins an existing project on the portal or creates a new project. Only faculty and senior technical staff can create projects. |
| 55 | 3. Creates a slice. |
| 56 | 4. Decides what resources she needs and which aggregates they will come from. |
| 57 | 5. Crafts a rspec that describes the resources she needs. She will typically use a GENI experimenter tool to craft this Rspec. |
| 58 | 6. Invokes the appropriate GENI AM API calls on the aggregates she wants to use; the request Rspec is sent to the aggregates using these AM API calls. |
| 59 | 7. Receives from the aggregates a manifest Rspec that includes information she will need to use the resources allocated to her slice. |
| 60 | 8. Use the resources she got to run her experiment. She may log into the compute resources she got, install software and send traffic on the network links. |
| 61 | 9. When the experiment is done, she deletes the resources from her slice. |