Skip to content

Installation via Docker image

This Docker image is the free version of Flow.

All search features are enabled but some limitations are present:

  • Limited to 5,000 images in collection
  • Limited to 1 collection
  • No distributed search support

If you want to go pro check out our license plans.

Download & run Flow image

Download and start the free pre-configured Flow Docker.

Minimum system requirements
  • CPU: x86_64 / AMD64 architecture (ARM and PowerPC not supported); E.g. at least Intel Core 6th generation, but newer is better (faster indexing and searching); performance measurements made with Intel Core i9 @ 12th generation
  • Memory: 4GB
  • Disk space: 5GB
  • OS: Linux
docker run --rm -p8983:8983 pixolution/flow

The server port 8983 is exposed. The log is displayed in your terminal and Flow is up and running.

How to persist data

To persist your indexed data between Docker restarts, create a named volume:

docker volume create flow-volume
and then mount it when starting Flow:
docker run -v flow-volume:/var/solr -p8983:8983 pixolution/flow

Add example dataset

The following command uses our example json with about 1,500 pre-analyzed docs and adds them to my-collection of your local Flow instance.

Index example dataset.
curl -s 'https://docs.pixolution.io/assets/docs_analyzed.json' | \
curl --header 'Content-Type: application/json' \
--data @- 'http://localhost:8983/api/cores/my-collection/update?commit=true'

Play around!

After starting the Docker image and indexing the example dataset, you are ready to go:

Open Flow in browser

The HTML user interface is designed for quick testing and has limited functionality. Flow is designed to be used by API.

Next Steps