Migrating a monolith to Cloud-Native and the stumbling blocks
summary
AsgarLabs replatforms by stuffing WAR/EARs into an Open Liberty container and running it on OpenShift, then treats pods like VMs so a batch job pins a single node at a 99% node CPU spike and Opex rises. An Enterprise License Agreement forces a cloud choice, DB2 and legacy schedulers duplicate Kubernetes features (and an F5 load balancer becomes redundant). Real fixes require refactoring into microservices with the Strangler pattern, CI/CD pipelines, distributed tracing (Jaeger/APM), standardized logging, and executive-backed team rotation to succeed.
AI-written summary · about
chapters (14)
- 0:00Introduction and speaker intro
- 1:06Why migrate to cloud native
- 3:38Containerization misconceptions
- 7:30Re-platforming vs re-architecting
- 10:40Repackaging into microservices
- 13:26Change Control Board shift
- 16:33Automation and pipelines
- 18:41Microservices outage challenges
- 20:51Cultural shift in enterprises
- 24:02Team collaboration strategies
- 26:09Visibility and monitoring needs
- 27:12Cost and accounting impacts
- 28:34Support and community reliance
- 29:15Do you really need cloud native
related talks
25mMigrating a monolith to Cloud-Native and the stumbling blocks that you don’t know about.
38mThe Good, the Bad & the Ugly of Migrating Hundreds of Legacy Applications to Kubernetes
09mThe cloud native elephant in the room
26mLessons Learned Migrating an Existing Product to a Multi Tenant Cloud Native Environment
14mKeynote: The Big Leap: Conquering the Legacy System and Building a Scalable Cloud Infra...
48mThe Journey of a strongly coupled Monolith into the Cloud Land
45mHow we rebuild a legacy monolithic platform at CERN and still sleep at night
51mEscape from Legacy Hell: A Java App Cloud Transformation