BonFIRE Services

BonFIRE is founded on 4 pillars, purposefully built for testing and experimentation

Control
  • Controlled physical host for execution
  • Exclusive access to physical host
  • Custom VM instance types
  • Emulated packet-loss and latency
Advanced Features
  • Bandwidth on demand
  • VM migration
  • CO2 and energy metrics
  • Bursting to commercial clouds
  • Controlled contentious and malicious patterns
Observability
  • Known physical host for execution
  • Known physical host architecture
  • Infrastructure, VM and Application metrics
  • API to monitoring data
Ease of Use
  • REST API to all functions
  • Single sign-on
  • Portal GUI and API builder
  • Declarative experiment descriptor
  • 380 pages of user documentation

Managed experiment environment

BonFIRE gives you control of your experiment, which is treated as a concrete entity in BonFIRE to manage your resources. Not only can you define your entire infrastructure across all testbeds in a single experiment descriptor, BonFIRE offers many features that make your life as an experimenter easier; for example:
  • Saving compute disk images with your personal software stack, as well as storage resources.
  • Sharing saved compute and storage resources.
  • Sharing access to experiments with colleagues.
  • Repeating experiments and share experiment descriptions for others to set up.
  • Aggregated monitoring metrics at both resource level (e.g., CPU usage, packet delay, etc.) and application level for your VMs.
  • Aggregated monitoring metrics at infrastructure level at selected testbeds.

Ease of use

While BonFIRE gives you deep control of resources to configure, execute and manage your experiment, we also strive to make this as easy as possible to do.

There are several ways in which you can use the BonFIRE API, depending on your preference. For example,
  1. You can create your experiment in a step-by-step manner via the BonFIRE Portal (GUI).
  2. You can use a command line client tool, such as or Python tools or Restfully, to interact with BonFIRE.
  3. You can write a script for your experiment deployment, which can be automatically executed by, e.g., Restfully.
  4. You can write a BonFIRE experiment descriptor, based on JSON or OVF.
  5. You can also issue raw HTTP commands via cURL, which you can also script if you wish.
For more information about using the different client tools, please click [here], or refer to the more comprehensive user documentation.

What can be tested and how?

BonFIRE supports experiments exploring the interactions between novel service and network infrastructures. Three initial scenarios have been defined to highlight the general classes of experiment that can be supported by the facility. The scenarios include:

  • Extended cloud scenario: the extension of current cloud offerings towards a federated facility with heterogeneous virtualized resources and best-effort Internet interconnectivity.
  • Cloud with emulated network implications: a controlled network environment by providing an experimental network emulation platform to service developers, where topology configuration and resource usage is under full control of the experimental researcher.
  • Extended cloud with complex physical network implications: investigates federation mechanisms for an experimental cloud system that interconnects individual BonFIRE sites with Federica, Open Cirrus and Panlab.
Please follow the links for each scenario for more details. For more details about the existing experiments on BonFIRE, click [here].