’s journey to cloud-native architecture
summary
Atlassian migrates more than 100,000 Jira and Confluence instances to a multi-tenant AWS backend, forking the monoliths and rearchitecting them into stateless apps backed by a common platform. The effort produces over 1,000+ microservices, shifts deploy cadence from weekly to multiple times daily, and cuts hotfix time from ~24 hours to ~30 minutes. Teams build tooling like a microscope service catalog and an automated service quest for launch readiness, and use Trello-driven SWAT work to finish the final 500 snowflake migrations.
AI-written summary · about
chapters (17)
- 0:00Introduction to Atlassian
- 1:34The monolithic past
- 2:57Project Vertigo starts
- 5:44Cultural shift
- 7:03The migration process
- 8:55Benefits of cloud native
- 11:32Key results
- 15:06Current state
- 17:36Setting clear goals
- 21:56Building internal tools
- 26:12Service quest readiness
- 28:21Organizational culture
- 31:42Balancing speed and trust
- 33:53Avoiding workarounds
- 35:17Migrating the last 10%
- 36:42Checks and balances
- 38:19Q&A session
related talks
33mGetting Cloud Migration Right
22mDecomposition and beyond: frontend engineering at scale
29mMulti Region Cloud Architecture for Data Residency Compliance
39mLessons Learned Migrating an Existing Product to a Multi Tenant Cloud Native Environment
38mThe Good, the Bad & the Ugly of Migrating Hundreds of Legacy Applications to Kubernetes
23mThe Nine Journey: Cloud Native Transformation
45mHow we rebuild a legacy monolithic platform at CERN and still sleep at night
25mMass migrations to the cloud