MACI: Seamless Execution and Analysis of Extensive Network Experiments

MACI is a framework for the management, scalable execution, and interactive analysis of extensive network evaluations consisting of a large number of experiments.

Figure 1: Experiment-driven research process.

1. Demo

2. Getting Started

MACI is self-explaining and provides a web interface at http://<SERVER_IP>:63658 (e.g., http://localhost:63658). Use one of the following options to get started:

  • Alternative 1: Get the source and follow the instructions in the readme, which basically boils down to:
    git clone
    cd maci
  • Alternative 2: A full MACI setup can be launched using docker-compose as described here.
  • Alternative 3: A ready to use AWS AMI is available in the AWS region Frankfurt. Follow these steps to start your AWS AMI:
    • Login at your AWS Console
    • In the AWS Console, select the EC2 service and select Launch Instance
    • Select Community AMIs, search for maci-v1, and click select
    • Follow the instructions and configure your instance
    • Login at your instance and run ./ You may want to use ssh port forwarding to access the web interface, e.g., with the following command ssh -L 63658:localhost:63658 8888:localhost:8888 <host>.

3. Overview Paper

A detailed paper about MACI was published as part CoNEXT'18 and is available here. If you use MACI in scientific papers, please use the following citation.

Don't Repeat Yourself: Seamless Execution and Analysis of Extensive Network Experiments

by Alexander Frömmgen, Denny Stohr, Boris Koldehofe and Amr Rizk. In: Proceedings of the 14th International Conference on emerging Networking EXperiments and Technologies (CoNEXT'18)

4. Research Papers that used MACI

The following papers span over different topics such as video streaming, Multipath TCP and Topology Pattern Matching.

Please contact us if you want your publication to be added to this list.


This work has been funded by MAKI to make the Internet more adaptive.

This work was supported by the AWS Cloud Credits for Research program


Contact Alexander Frömmgen or Denny Stohr for any comments and questions.