• [+] expand all

Deploy to Docker

Deploy Vespa applications to Docker containers.

Install pyvespa

!pip install pyvespa

Define your application package

This tutorial assumes that a Vespa application package was defined and stored in the variable app_package. To illustrate this tutorial, we will use a basic question answering app from our gallery.

from vespa.gallery import QuestionAnswering

app_package = QuestionAnswering()

Docker requirement

This guide illustrate how to deploy a Vespa application to a Docker container in your local machine. It is required to have Docker installed in the machine you are running this tutorial from. For that reason we cannot run this tutorial in Google Colab as Docker is not available on their standard runtime machines.

Deploy to a Docker container

Create a VespaDocker instance based on the application package. Set the environment variable WORK_DIR to the absolute path of the desired working directory.

import os
from vespa.deployment import VespaDocker

disk_folder = os.path.join(os.getenv("WORK_DIR"), "sample_application")
vespa_docker = VespaDocker(port=8089, disk_folder=disk_folder)

Call the deploy method. Behind the scenes, pyvespa will write the Vespa config files and store them in the disk_folder, it will then run a Vespa engine Docker container and deploy those config files in the container.

app = vespa_docker.deploy(
    application_package = app_package,

That is it, you can now interact with your deployed application through the app instance.

Clean up environment

from shutil import rmtree

rmtree(disk_folder, ignore_errors=True)