Quickstart

Terminology Primer

FIREWHEEL has some specific terminology, which is useful to understand prior to running any Emulytics experiments:

• An Experiment is the coordinated realization of your topology (e.g. live VMs) and the scheduled actions which occur within that realization. There are a number of reasons for running an experiment, so the outputs from the experiment differ depending on the use-case (e.g. live exercise vs software analysis). See Cyber Experimentation for more details.

• In FIREWHEEL, an experiment is created via one or more Model Components. A Model Component (MC) is a colocated set of related code, VM Resources, and/or VM images. That is, model components are the building blocks of experiments. See Model Components for a more detailed description.

• VM Resources (VMRs) can be any software or files, that runs on or otherwise interacts with running virtual machines in order to perform a desired function on them (e.g. install software, configure settings, collect data, etc.). VM Resources are run via schedule, which users define in their Model Components. The VMR schedule is executed via the VM Resource Manager.

Installing FIREWHEEL

While at it’s core, FIREWHEEL is a Python package to realize cluster-based communications and emulation, the installation is more advanced than typical packages. Because of the necessary configuration options, we recommend following the full installation guide Installing FIREWHEEL. This quick start guide assumes that FIREWHEEL has been installed and is ready to be used.

Getting Started

We highly recommend reviewing our Tutorials for using FIREWHEEL. After completing these tutorials, users should understand how to create and run a new experiment.

• Running the Router Tree Topology - Learn how to run an example topology with FIREWHEEL.

• Simple Server Tutorials - Learn how to create a relatively simple experiment in which you will start a simple web server, grab a web page, and analyze the resulting data.

When building a new experiment, it’s important to have realistic expectations about the complexity of building off-line Emulytics experiments. While users have full access to the FIREWHEEL Model Component library, they will likely need to code their own topologies and determine how to install/configure necessary experiment-related software in an offline-environment. We recommend answering these questions before beginning a new experiment:

• What the topology should look like?

  • What VMs are needed?

  • How they should the VMs be connected?

• What VM resources are needed?

  • What software should be installed on the VMs and can it be installed in an offline-environment?

  • How should the software be configured?

  • How can the software installation/configuration be scripted/automated?

• What data (if any) should be collected from the experiment?

  • How can that data be extracted from the experiment?

Helpful Commands

All the FIREWHEEL CLI commands are documented in Command Line Interface, however, here are a few of the more used commands.

Note

All the below commands should be prefaced with firewheel (e.g. firewheel experiment).

• experiment - The command how to execute model components and, thereby, start an experiment.

• vm list and vm mix - These commands help show the state of the currently running experiment. vm list has a number of options to show information about each VM.

• restart - To end a running experiment and reset the testbed for a new experiment, run this command.