Migrating legacy monolith applications to modern microservices architectures presents significant challenges that often involve a cycle of trial and error.
Modernization improves security, reliability, and scalability, and can be approached using strategies like rehosting, replatforming, or refactoring.
Konveyor is a CNCF open source project that analyzes source code and artifacts using a customizable rules engine to identify migration issues.
Konveyor AI enhances static analysis by connecting to large language models (LLMs) to automatically generate complete code solutions for migration issues.
A live demonstration shows how to set up the local environment, including a Postgres database and LLM server, and configure an analysis in VS Code.
Konveyor AI uses Retrieval-Augmented Generation (RAG) to overcome LLM context limits and improve accuracy by leveraging a database of previously solved migration patterns.
The analysis report is reviewed in VS Code, where AI-generated code snippets for fixing issues like namespace changes and API refactoring can be accepted with a single click.
After applying all AI-suggested fixes, the legacy application is successfully migrated and runs on the modern Quarkus stack, complete with its dependencies.
Unlike developer-centric tools like Copilot, Konveyor AI is designed for organization-wide analysis across multiple repositories, creating a shared knowledge base of migration solutions.
41:17Syed M Shaaf
26:55Shaaf
29:18Daniel Oh & Kevin Dubois
32:02Kevin Dubois
31:02Daniel Oh
28:21Daniel Oh
31:39Rainer Stropek
15:26Gregor Schumacher, Sujay Joshy & Marcel Gocke
Jobs that call for the skills explored in this talk.
Amdocs
Kontich, Belgium
webLyzard
Vienna, Austria
codemanufaktur GmbH