Designing software in the cloud
summary
A legacy monolith migrates to Azure as many App Services and creates a fragile distributed monolith that fails after eight days; the migration cost spikes to 40-50 App Services on separate plans. The fix refactors into queue-driven flows using App Service APIs, storage queue between Event Grid and Functions to avoid lost events, Azure Functions with SignalR bindings for async client notifications, and selective Service Bus when features are required — this reduces coupling, improves replayability, and controls cost.
AI-written summary · about
chapters (4)
- 0:00Start
- 9:48The innards of a (micro)service
- 14:38Which Azure Services to use?
- 22:59Bad practice - Bloated events
related talks
44mMigrating Legacy Applications to the Cloud
01h 08mBuild Scalable Web Apps with .NET and Azure
42mDo proper monoliths before you do microservices!
33mChopping the monolith
31mEvent-Driven Architecture - A successful strategy for a journey to the cloud
36mReinventing the Way of Building a Core Software Product in a 50 Year Old IT Co
23mMigration of a containerized legacy application to the cloud
35mBuilding Event Driven Integration Architectures