Knee Deep in Microservices
summary
Argues microservices make sense at any team size but impose real costs: test/production parity, networking, service discovery, monitoring and culture. Recommends packaging with Docker chainsaw, running under an orchestrator (Mesos/DC/OS or Kubernetes) to get virtual networking, per-container IPs, service discovery and load balancing; uses Prometheus/cAdvisor/Datadog/Stackdriver for monitoring and autoscaling examples. Shows manifest-driven deployment as the turning point that shifts operational burdens into a platform and enables independent CI/CD for each service.
AI-written summary · about
chapters (20)
- 0:00Intro
- 0:25ADAM SANDOR
- 1:55DEMONS OF MICFOSERVICES
- 2:50CLOUD NATIVE WEAPONRY
- 3:37MICROSERVICES
- 8:36DOCKER
- 10:00CONTINOUS DELIVERY
- 12:20ENTERING SCHEDULING
- 13:02RESROUCE ALLOCATION
- 16:25SCHEDULERS
- 17:31NETWORKING
- 22:21SERVICE ABSTRACTION
- 23:32MICROSERVICE PLATFORMS
- 24:37OBSERVABILITY
- 26:11CN MONITORING Prometheus
- 29:21LOG AGGREGATION
- 29:51ALERTING
- 30:28HELL IS OTHER PEOPLE
- 32:29MANIFEST FILES
- 36:17THE CLOUD-NATIVE ATTITUDE
related talks
49mMicroservices: An unexpected Journey
01h 18mIn Defense of Microservices
01h 03mThe Magic of Small Things - 10 Years of Microservices
55mMicroservices, Where Did It All Go Wrong
57mEmbracing Microservices
45mContinuously delivering containerized, highly available microservices
50mEmbracing Microservices
27mMicroservices Are ‘Easy’, Dependencies Are Hard