10 parts · 12 chapters
The Production Stack
No runtime runs alone in production. Django sits behind Gunicorn and Nginx with Celery beside it; Rails runs on Puma with Sidekiq; Go ships as one binary behind Envoy; Spring Boot talks to Eureka and a config server. This course is about the things every language is paired with, why each one is there, and in what order companies add them.
Ten parts: the anatomy of a production request; reverse proxies; app servers and process managers; connection poolers; caches and queues as standard equipment; background workers; service discovery and API gateways; observability agents; pairing tables for every major runtime; and how a company's stack grows from one box to a platform.
the requestDNS, CDN, load balancer, gateway, proxy, app, cache, database, queue: one request through all of it.
proxies and app serversNginx, Caddy, HAProxy, Envoy; PM2, Gunicorn, Uvicorn, Puma, Tomcat, Netty.
poolers, caches, queuesPgBouncer, ProxySQL, Redis, Memcached, RabbitMQ, SQS, Kafka, NATS.
workersBullMQ, Celery, Sidekiq, Solid Queue, asynq, Spring Batch.
discovery and gatewaysConsul, Eureka, DNS, Kong, cloud API gateways.
pairing tablesThe usual stack for Node, Python, Go, Rust, Rails, Java, .NET and Elixir at 1, 10 and 100 services.
00
Anatomy of a Production Request
Every box, in order
1 ch · ~8 min01Reverse Proxies and Web Servers
Nginx, Caddy, HAProxy, Envoy
1 ch · ~8 min02App Servers and Process Managers
Process, thread and event-loop servers · Process managers versus orchestrators
2 ch · ~12 min03Connection Poolers
Poolers per runtime
1 ch · ~8 min04Caches and Queues as Standard Equipment
Which one, and when
1 ch · ~8 min05Background Workers
One pattern, every language
1 ch · ~8 min06Service Discovery and API Gateways
Finding healthy instances · API gateways
2 ch · ~12 min07Observability Agents
Collectors, shippers and per-runtime instrumentation
1 ch · ~8 min08Pairing Tables by Runtime
What each runtime is paired with
1 ch · ~8 min09How a Company's Stack Grows
Stage by stage
1 ch · ~8 minBuilt on Infra, Cloud and SREAssumes Cloud Engineering for networking, Infrastructure for Kubernetes and CI/CD, and SRE for reliability. Courses 10 to 12 go deeper on configuration, deployment and service meshes.