Docker · Containers in practice
When would you use Compose versus Kubernetes?
Short answer
Compose runs a stack on a single host with minimal operational overhead. Kubernetes schedules across nodes and provides self-healing, scaling and rollout primitives, at the cost of complexity.
Compose is a process manager for containers on one machine. Kubernetes is a distributed control plane that places workloads on a cluster, reconciles desired state, and exposes services, configuration and rollout mechanisms. They solve different scopes rather than competing directly.
Decision factors
- Scale: one host with Compose, many nodes with Kubernetes
- Availability: Compose has no scheduler or rescheduling across hosts
- Rollouts: Kubernetes supports rolling updates and rollbacks natively
- Operations: Compose is simple to run; Kubernetes needs a team and tooling
- Portability: Compose is convenient locally even when production uses Kubernetes
A common setup uses Compose for local development and Kubernetes in production, keeping the container images and configuration contract aligned. Tools can translate Compose files, but generated manifests still need review for probes, resources and storage.
If a workload runs comfortably on one host and downtime is acceptable, Compose plus a reverse proxy is a legitimate production choice. Do not adopt Kubernetes only for resume value; the operational cost is real.
Follow-ups you should expect
Interviewers use these to check whether the first answer was understanding or memorisation.
- 1What does Kubernetes provide that Compose does not?
- 2Can you use both in the same workflow?
- 3What is the operational cost of running Kubernetes?
- 4How would you migrate a Compose stack to Kubernetes?
Say it in the interview
Compose manages containers on a single host and is ideal for local development and small deployments. Kubernetes adds scheduling across nodes, self-healing, rolling updates and service discovery, but it needs real operational investment. I choose based on scale and availability requirements, and it is normal to use Compose locally while production runs on Kubernetes.
Use this as a model, not a script. Saying it in your own words is what makes it survive follow-up pressure.
Related questions
Reading is step one
Now say it out loud.
Understanding an answer and delivering it under pressure are different skills. Rehearse this topic in a live Docker mock interview and get scored feedback on depth, clarity and confidence.