How to Escape The Distributed Monolith

on YouTube ↗59m

" summary

Companies with ~700 engineers and ~52 teams find RPC-based microservices become a distributed monolith because services use HTTP/gRPC and service meshes as proxies, creating temporal coupling and brittle releases. The practical alternative is to publish events (Kafka, RabbitMQ, SNS/SQS), use queues for guaranteed delivery, build sagas/conversations for orchestration, and replicate reference data (event/state-transfer or caches) so bookings, payments, housekeeping and channel management can proceed independently and recover asynchronously.

AI-written summary · about

" chapters (12)

1. 0:00Start
2. 3:09Agenda
3. 7:28The Inverse Conway's Law Maneuver
4. 9:10Business Capabilities
5. 10:33Business Capability Mapping
6. 11:32Foundational Capabilities
7. 25:21Behavioral Coupling
8. 34:09Processes in Mature Businesses To Deal with Issues of Lack of Availability
9. 36:02Guarantee Delivery
10. 38:18Eventual Consistency
11. 46:26Managing Error Flows
12. 46:49Data on the Inside versus Data on the Outside

" related talks

• 01h 03mMicroservices vs The Distributed MonolithChris Riley · PHP UK Conference 2024
• 53mDon’t Build a Distributed Monolith: How to Avoid Doing Microservices WrongJonathan "J." Tower · NDC Porto 2023
• 48mMicroservices vs The Distributed MonolithChris Riley · Dutch PHP Conference 2018
• 38mMonolith vs. MicroservicesRafa Paradela · Scala Days 2023 Seattle
• 43mMicroMonolith - Top anti-patterns of distributed systemsMichal Franc · NDC Oslo 2016
• 48mMicroMonolithMichal Franc · Codemotion Amsterdam 2017
• 43mHow to Avoid Doing Microservices Completely WrongJonathan "J." Tower · Update Conference Krakow 2025
• 42mQ&A - Monolith to MicroservicesSam Newman, Sven Johann · GOTO Oslo 2020