What is the difference between Docker Swarm and Kubernetes?

Both orchestrate containers across multiple hosts. Docker Swarm is simpler to set up and sufficient for many use cases. Kubernetes is far more feature-rich (RBAC, custom resources, operator pattern) and is the industry standard for complex production workloads.

What is a liveness probe vs readiness probe?

Liveness probe: checks if the container is alive — failure triggers a restart. Readiness probe: checks if the container is ready to receive traffic — failure removes the pod from the load balancer. Both are critical for zero-downtime deployments.

What is the difference between a Deployment and a StatefulSet?

Deployments manage stateless pods interchangeably — any replica can serve any request. StatefulSets manage pods with stable network IDs and persistent storage, required for databases and other stateful workloads where identity matters.

Container Orchestration Simulator

DevTools Surf

Example: Container Orchestration Simulator

A worked example, rendered from real sample data. Sign in to run the tool on your own input.

Input

strategy=rolling
replicas=6
maxSurge=25%
maxUnavailable=25%
readinessSeconds=15
terminationGracePeriodSeconds=30
name=api
namespace=production
image=ghcr.io/acme/api:1.4.2

Output

═══ What this tool did ═══
✗ No cluster was contacted. Nothing here is read from a running Kubernetes API server: there are no pod counts, no CPU numbers and no memory numbers from any real workload. This page runs entirely on the server that rendered it and makes no outbound network calls.
✓ What it does compute is arithmetic on the settings you gave: how many pods the Deployment controller will create and delete at each step, how many stay available, and what the manifest for that has to say.

═══ Settings read from your input ═══
Strategy: rolling
Desired replicas: 6
Deployment: production/api
New image: ghcr.io/acme/api:1.4.2
Readiness delay assumed per pod: 15s
Termination grace period assumed per pod: 30s
maxSurge: 25% → 2 extra pods (percentages round UP for surge)
maxUnavailable: 25% → 1 pod (percentages round DOWN for unavailable)
ℹ The two timings are the only assumptions in this plan and they come from your input, not from a measurement. Every pod count below follows from them by arithmetic.

═══ Rollout steps ═══
t+    action                                old  new  available
----  ------------------------------------  ---  ---  ---------
0s    steady state before the rollout       6    0    6
0s    create 2 new pods                     6    2    6
15s   2 new pods pass the readiness probe   6    2    8
45s   terminate 3 old pods                  3    2    5
45s   create 3 new pods                     3    5    5
60s   3 new pods pass the readiness probe   3    5    8
90s   terminate 3 old pods                  0    5    5
90s   create 1 new pod
…

About Container Orchestration Simulator

Container Orchestration Simulator preview - Developer Utilities tool

Model a Kubernetes rollout: exact pod counts per step for rolling, blue-green, canary and recreate, plus the manifest and kubectl commands. Part of the DevTools Surf developer suite. Browse more tools in the Developer Utilities collection.

Use Cases

Practice Kubernetes deployment scenarios without requiring a cluster
Simulate pod scheduling and eviction during node failure events
Test scaling policies before deploying to production
Demonstrate orchestration concepts during team education sessions

Tips

Configure the replica count and resource limits (CPU, memory) per service before running the simulation — these drive scheduling decisions
Use the 'node failure' event to simulate what happens when one of three nodes goes down — watch which pods reschedule and where
The failover timeline shows the expected recovery duration based on configured liveness/readiness probe settings

Fun Facts

Kubernetes was open-sourced by Google in June 2014, based on Google's internal Borg system that had been running production workloads since 2003. Google has been running containerized workloads at scale for over 20 years.
As of 2023, over 5.6 million developers use Kubernetes according to the CNCF Annual Survey. It runs in over 96% of Fortune 100 companies and processes billions of container instances daily across hyperscaler infrastructure.
The name 'Kubernetes' comes from Greek for 'helmsman' or 'pilot' — the same root as 'cybernetics' (control systems theory). Docker containers were named after shipping containers; Kubernetes continues the nautical metaphor.

FAQ

What is the difference between Docker Swarm and Kubernetes?: Both orchestrate containers across multiple hosts. Docker Swarm is simpler to set up and sufficient for many use cases. Kubernetes is far more feature-rich (RBAC, custom resources, operator pattern) and is the industry standard for complex production workloads.
What is a liveness probe vs readiness probe?: Liveness probe: checks if the container is alive — failure triggers a restart. Readiness probe: checks if the container is ready to receive traffic — failure removes the pod from the load balancer. Both are critical for zero-downtime deployments.
What is the difference between a Deployment and a StatefulSet?: Deployments manage stateless pods interchangeably — any replica can serve any request. StatefulSets manage pods with stable network IDs and persistent storage, required for databases and other stateful workloads where identity matters.

Related Developer Utilities Tools

Collection JSON → cURL Git Diff → HTML Regex Visualizer Makefile Explainer Shell Script Linter GitHub Actions Visualizer HAR File Viewer API Response Mocker