4.2) I&M Use Cases for Experiments, and Support for Experimenters

1) Goals

From Sec. 2 of the GENI I&M Architecture document:

To conduct useful and repeatable experiments, GENI experimenters require extensive and reliable instrumentation and measurement capabilities to gather, analyze, present and archive Measurement Data (MD).

Remove the burden on researcher to become a system and network measurement infrastructure expert so that researcher can better focus on the science in the experiments

Work towards Max Ott's vision for experiment support

​slides

Define use cases for a wide range of experiments: complex (by skilled experimenters) to simple (by novice experimenter, i.e., a student)

2) Tasks

Provide a concise but complete definition of I&M Use Cases for Experiments

Identify the support that should be available to experimenters, i.e., on-line instructions, examples and help desks.

Update the ​GENI I&M Architecture document:

Sec. 3.1. I&M Use cases for Experimenters
Sec. 4.2.1 Typical Arrangements of I&M Services: For Experimenter Gathering MD from their Slice
Sec. 4.3.1 Type 1 I&M Service: Service Contained within a Slice

Use as guidance in the design of GENI I&M tools, particularly for the GEMINI and GIMI projects

3) Team

CO-LEAD Jim Griffioen (U Kentucky)
Hussamuddin Nasir (U Kentucky)
CO-LEAD Mike Zink (UMass Amherst)
David Irwin (UMass Amherst)
Max Ott (NICTA)
Prasad Calyam (Ohio Super Center)
Harry Mussman (GPO)

4) Meetings

(organized calls or meetings before GEC13?)

Review conclusions in pre-meeting at GEC13

Review with working team at GEC13

Review with experimenters at GEC13? how?

5) Issues

1) Can an Operator (e.g., GMOC) ever see MD owned by an Experimenter, e.g., to verify acceptable use of GENI resources or to troubleshoot problems in GENI infrastructure? If so, how?

6 Use Cases

1) Baseline experiment

Outline:

1) Setup and run experiment using Experiment Control Tools and Experiment Orchestration Service

Include resources and software for instrumentation and measurement services in the slice

2) Setup and run measurement services within slice, using persistent services as necessary.

Option: get measurement data from persistent MP.

3) Gather MD into MAP service, to analyze and.or visualize, during or after experiment; formulate MDOD.

4) Archive MD with MDOD.

5) Share archived MD with others, per policy included within MDOD.

6) Pull MD out of archive, to analyze and/or visualize.

Example: INSTOOLS

2) Where MC may be disconnected from MPs from time to time

Example: WiMAX (or other wireless) experiment

3) Experiment where measurements may drive other functions in the experiment

Example: adjust protocol parameters based on LEARN/BEN layer 0,1 or 2 measurements

4) Sensor measurements

Example: radars and power monitors
Where measurements are at the heart of the experiment

5) Workflow measurements

Example: measure DiCloud workflow

7 Measurement Parameters

1) Baseline host measurements

A typical experimenter's slice includes several nodes or hosts, with connecting links. Baseline host measurements include a set of measurements for each node or host.

From INSTOOLS, this is: ​baseline measurements of each node (host)

2) Custom slice measurements

An experimenter must be able to define custome measurements in their slice, i.e., be able to create custom MPs

8) I&M Support for Experimenters

Support for Experimenters:

On-line instructions
Examples
Help desks
How?