Watch the demo: https://youtu.be/tmGSqK3BahU Konveyor Move2Kube helps automate your migration from various platforms such as cloud foundry, docker swarm, VM based applications, or from your custom framework to Kubernetes. With the latest release of Move2Kube, v0.3.0, it has acquired extensive customization capabilities to help create the target artifacts as per your organization requirements. In this session, the Move2Kube team will demonstrate the migration of a typical application to run on Kubernetes. It will be followed by a demonstration of customization capabilities of Move2Kube, such as customizing the generated Dockerfile, the directory structure, the helm chart and custom artifacts. Website: https://move2kube.konveyor.io/ GitHub repo: https://github.com/konveyor/move2kube/

1. Automate the creation/transformation of all your
Infrastructure as Code artifacts
Konveyor Move2Kube
1

2. 2
Accelerate your journey to Kubernetes
with the Konveyor Community
A community of people passionate about
helping others modernize and migrate
their applications to the hybrid cloud by
building tools and best practices on
how to break down monoliths, adopt
containers, and embrace Kubernetes.
www.konveyor.io

3. Watch The Presentation
https://youtu.be/tmGSqK3BahU

4. 4
What we’ll
discuss
today...
▸ Konveyor Move2Kube -
Introduction
▸ Capabilities
▸ Customizations

5. Konveyor Move2Kube
Discover Transform Parameterize &
Customize
Move2Kube allows you to create all your Infrastructure as Code artifacts as per your organizational requirements.
It allows integrated discovery, containerization, transformation, parameterization and customization.
…
Language stacks
Any specific Kubernetes
flavor/version
Dockerfile
Containerize
https://move2kube.konveyor.io/
Kustomize Templates
Custom
Application
Custom Artifacts

6. Konveyor Aided Migration
Apps
Services
Discovery
Discovery
Pick Representative
apps
Pick Service types
Operator
Create migration templates
per org requirements
Create migration templates
per org requirements
Perform human-in-the-loop
automated migration
Perform human-in-the-loop
automated migration

7. Usage modes
Releases
bash <(curl https://raw.githubusercontent.com/konveyor/move2kube/main/scripts/install.sh)
https://github.com/konveyor/move2kube
Command line tool Web Interface
Operator
brew tap konveyor/move2kube
brew install move2kube

8. Target-artifacts
$ move2kube transform –s src/
One step usage
Edit CI/CD Pipeline
Deploy
Next Steps…

9. Target-artifacts
Optional:
$ move2kube collect
Step 1:
$ move2kube plan –s src/
Step 2:
$ move2kube transform
Collect crawls metadata about the source and
target runtime environments such as:
• Supported object kinds in cluster
• Apps running in cloud foundry instance
• Meta-information from local docker
images.
Inputs: The terminal context should have cf
and kubectl logged in.
Outputs: Data from runtime instances as files.
Plan generates a plan file containing a
transformation proposal (including
containerization options) for all services
discovered from various sources.
Inputs: src – Directory containing source code
and collected artifact files.
Outputs: Plan file
Transform transforms the input source artifacts.
as per the generated plan, into target artifacts
containing:
Inputs: Plan file and src.
Outputs:
• Scripts for containerization.
• Helm chart, Kustomize, Openshift
templates, docker-compose.
Move2Kube
Involved Usage
Scrape from source and target
runtime environments
Analyze code & collected artifacts and
correlate Optimize and Translate
Edit CI/CD Pipeline
Deploy
Next Steps…

10. Move2Kube Transformer Framework
Nodejs
Java
..
Source
Dockerfile
Parser
..
K8s yaml
generator
..
Helm chart
generator
..
Destination
Artifacts

11. Transformer
Transformer
Transformer
Source
Plan Phase
Transform Phase
Artifact
Artifact
Artifact
Path Mapping

12. Customizing Transformers
https://github.com/konveyor/move2kube-transformers
Customizing in-built
transformers
Each transformer exposes configurations which can
be used to customize the transformer
Custom transformers in
starlark
A completely functional transformer can be written
in starlark (python like)
Custom transformers as
container
A completely functional transformer can be written
in any language, and can be packaged as a
container

13. Sample Usecases
…
Language stacks
https://move2kube.konveyor.io/tutorials
…
Language stacks
Custom Artifacts

14. Web UI – with Authorization and multi-tenancy
docker run --rm -it -p 8080:8080 -v "${PWD}/move2kube-api-data:/move2kube-api/data" quay.io/konveyor/move2kube-ui:latest

15. Move2Kube Case Study
For ~100 container scenario.
Based on extrapolation from PoC experience on a
limited subset of the migration process, we will further
evolve this from actual delivery data
Tasks Manual (est.) Move2Kube
Discover relevant assets 3 days 3 hours
Direct artifact translations 14 days 1 day
Complex artifact translations 32 days 4 days
Exploit OCP features + best practices 3 days 3 hours
Customize for a deployment 4 days 30 mins
Rightsize initial config 5 days 1 day
61 days 7 days
~9x
reduction
Speed up modernization (Move2Kube) Istio
Application and middleware
services on Docker Swarm
Application and middleware
service on Red Hat Openshift

16. Contributions are welcome!
Head over and submit PRs
https://github.com/konveyor/move2kube-ui
https://github.com/konveyor/move2kube-tests
https://github.com/konveyor/move2kube
https://github.com/konveyor/move2kube-api
https://github.com/konveyor/move2kube-operator
https://github.com/konveyor/homebrew-move2kube

17. Join the Konveyor Community
www.konveyor.io

18. Thank you!
18
Ashok Pon Kumar
ashokponkumar@in.ibm.com
Padmanabha Venkatagiri Seshadri
seshapad@in.ibm.com
Harikrishnan Balagopal
harikrishnan.balagopal@ibm.com
Akash Nayak
Akash.Nayak1@ibm.com
Konveyor Move2Kube Maintainers
IBM Research