Skip to main content

SAND: System Architecture as a Network of Dependencies

Project description


Python code and notebooks to model System Architecture as a Network of Dependencies

SAND uses Python and Jupyter Notebooks to explore applications of representing system architecture as a directed graph, or network, of engineered artifacts and their relationships to one another.

Engineered artifacts are vertices in the graph. For a software library, the artifacts are functions and the dependencies are function calls. For RESTful microservices, an artifact is a service and the dependencies are API calls.

Directed edges represent the dependencies and their transpose, impact.

Imagine we have two microservices, A and B.

If B calls A, then B has a dependency on A. The creator of A might not know that B is a client, so the dependency relationship is directed.

The transpose of this relationship is that A impacts or influences B: Non backwards-compatible changes in A’s interface that B calls can break B. Changes in B do not impact A, so once again, the edge is directed.

This simple model proves to be extremely powerful in describing arbitrarily complicated system architectures. The SAND library and accompanying Jupyter Notebooks provide working examples of visualization and analysis.


pip install sand

You might also want to clone this git repo to follow along with the examples below:

git clone
cd sand

Getting Started

SAND is documented with a series of Jupyter Notebooks:

Running in Docker

You can run these notebooks via Docker to experiment. Assuming you have a docker-machine running and you’ve cloned the sand repo:

git clone
docker pull testedminds/sand
make docker-docs
# And after the container starts...
make docker-open

When the notebook opens in your browser, you will see the Notebook Dashboard, which will show a list of the notebooks and data in the docs directory.

These commands translate to:

docker run -d -p 80:8888 -v `pwd`/docs:/opt/sand --rm --name sand testedminds/sand:latest \
                jupyter notebook --allow-root --ip --no-browser --NotebookApp.token=''


This is a useful technique to quickly explore network data anywhere on your local filesystem.

Running locally

To run the notebooks locally without Docker:

pip install -r requirements.txt
cd docs
jupyter notebook
  • Install Cytoscape to run the optional Cytoscape examples. Start Cytoscape up and close the welcome screen. You probably want to check “Don’t show again” in the lower left.

Learn More

See a presentation from Bobby Norton at Windy City GraphDB for a more detailed introduction to the concept.

The Notebooks leverage Cytoscape’s RESTful API and python-igraph.


Copyright © Bobby Norton and Tested Minds, LLC.

Released under the Apache License, Version 2.0

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

sand-0.2.1.tar.gz (11.6 kB view hashes)

Uploaded source

Built Distribution

sand-0.2.1-py3-none-any.whl (15.7 kB view hashes)

Uploaded py3

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Huawei Huawei PSF Sponsor Microsoft Microsoft PSF Sponsor NVIDIA NVIDIA PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page