devinrsmith / deephaven-server-docker

Deephaven Server Docker

deephaven-server-docker

This repository produces Deephaven server Docker images with the deephaven-core releases.

Quickstart

To get started quickly, simply run:

docker run \
    --rm \
    --name deephaven \
    -p 10000:10000 \
    ghcr.io/deephaven/server:0.20.0

This will start the server, and the web UI will be available at http://localhost:10000.

Images

The following server images are currently being produced:

• ghcr.io/deephaven/server:0.20.0
• ghcr.io/deephaven/server-slim:0.20.0
• ghcr.io/deephaven/server-all-ai:0.20.0
• ghcr.io/deephaven/server-nltk:0.20.0
• ghcr.io/deephaven/server-pytorch:0.20.0
• ghcr.io/deephaven/server-sklearn:0.20.0
• ghcr.io/deephaven/server-tensorflow:0.20.0

Linux

Images are produced for the linux/amd64 and linux/arm64 platforms. The images are based off of the ubuntu Docker image.

Mac

Both the Intel and M1 architectures are supported with the Linux images.

Windows

The Linux images can be used with the Windows Subsystem for Linux. Windows native images are not currently being produced, but may be produced in the future.

Build

CI

The images are automatically built and deployed by GitHub Actions CI, see release-ci.yml.

Local

For a default, local-only build on your system's platform, run:

# Build the default server image:
docker buildx bake -f server.hcl

# Build a specific target image:
docker buildx bake -f server.hcl server-all-ai

See the various hcl files for parameterization options.

Resources

• Issues
• Deephaven Community Slack
• Deephaven Documentation