Container Guide

Running the Dataverse software in containers is quite different than in a standard installation.

Both approaches have pros and cons. These days, containers are very often used for development and testing, but there is an ever rising move toward running applications in the cloud using container technology.

NOTE: As the Institute for Quantitative Social Sciences (IQSS) at Harvard is running a standard, non-containerized installation, container support described in this guide is mostly created and maintained by the Dataverse community on a best-effort basis.

This guide is not about installation on technology like Docker Swarm, Kubernetes, Rancher or other solutions to run containers in production. There is the Dataverse on K8s project for this purpose, as mentioned in the Docker, Kubernetes, and Containers section of the Developer Guide.

This guide focuses on describing the container images managed from the main Dataverse repository (again: by the community, not IQSS), their features and limitations. Instructions on how to build the images yourself and how to develop and extend them further are provided.