Docker Compose

Definition

Docker Compose is a tool for defining and running multi-container Docker applications using a single YAML configuration file that specifies each service, its image or build context, networking, and volumes. A single command then builds and starts all defined services together, making it a standard way to run local development environments composed of multiple interdependent containers. A single command builds and starts all defined services together with a shared network so containers can reach each other by service name, and its functionality is now integrated directly into the Docker CLI as the `docker compose` subcommand, superseding the earlier standalone Python-based `docker-compose` tool.

Overview

Docker Compose was created to address a common limitation of using the Docker CLI directly: running an application made of several containers, such as a web server, a database, and a cache, requires manually starting each container with correctly configured networking and volume flags. Compose lets a developer describe all of these services declaratively in a `compose.yaml` file, specifying image sources, environment variables, port mappings, and dependencies between services in one place. Mechanically, running `docker compose up` reads that file, builds any services with a defined build context, creates a shared network so containers can reach each other by service name rather than by IP address, and starts everything according to the file's configuration; `docker compose down` tears the whole set back down and removes the associated network. This makes it straightforward to reproduce a consistent local environment across a team, since the entire multi-container setup lives in a version-controlled file rather than a set of manual setup instructions in a wiki. Compose sits below full orchestration platforms in scope: it targets a single host by default, whereas Kubernetes or Docker Swarm coordinate containers across many hosts with scheduling, self-healing, and rolling updates. Docker later integrated Compose functionality directly into the Docker CLI as the `docker compose` subcommand, superseding the earlier standalone `docker-compose` Python-based tool, which simplified installation and kept the tool aligned with Docker Engine's own release cycle. In practice, Compose is most commonly used for local development and testing environments, where a developer runs the full application stack, including its database and supporting services, with one command instead of manually orchestrating several `docker run` invocations. Compose files can also define profiles to selectively start subsets of services, override files to layer environment-specific configuration on top of a base file, and health checks to control startup ordering more precisely than simple dependency declarations alone. While some teams historically used Compose for small-scale production deployments, orchestration platforms are generally used for production-grade multi-host container orchestration once an application outgrows a single machine. This flexibility has made Compose a common entry point for developers learning containerized application architecture before moving to full orchestration platforms. Developers moving an application from local Compose-based development toward production often translate the same service definitions into Kubernetes manifests or Swarm stack files by hand, since Compose's single-host model does not extend automatically to a distributed, self-healing multi-host deployment without that additional translation step. This extra translation step is one reason many teams eventually adopt Kubernetes-native tooling that can generate manifests from Compose files as a starting point rather than a finished migration.

Key Features

• Declarative YAML definition of multi-container application services
• Single-command startup and teardown of an entire application stack
• Automatic shared networking between services by name
• Volume and environment variable configuration per service
• Support for build contexts alongside prebuilt images
• Profiles for selectively starting subsets of services
• Override files for layering environment-specific configuration
• Integrated directly into the Docker CLI as `docker compose`

Use Cases

Alternatives

Frequently Asked Questions

Related in AI Tools