← कोर्सच्या मुख्य पानाकडे परत

🗓️ अभ्यास योजना — 13 धडे, 4 आठवडे

~3 sessions a week: read the lesson (~15 min) + run and change its section of scale/demo.py (~25 min). Tick lessons off — progress is saved in this browser.

🐢 स्थिर: आठवड्याला 3 सत्रे → 4 आठवड्यांत पूर्ण.
🐇 वेगवान: प्रत्येक भागासाठी एक weekend → 3 weekends मध्ये पूर्ण.
💰 Cost: zero. Python 3 on your laptop; no cloud account.
0 / 13

WEEK 1 ⏱️ Up or out, and count first

Why scaling, measuring load, CloudFront + S3. Lab: change the per-server number; add a slow request and watch p99; try TTL 0, 1, 10, 60.

🏁 Checkpoint: you can turn a traffic number into servers, and explain why TTL 10 beats TTL 1 by a mile.

WEEK 2 ⚡ The edge and the counters

Edge caching, stateless servers, Auto Scaling. Lab: add an origin shield; move Katrina's session to Redis; change the warm-up to 1 and 4 minutes.

🏁 Checkpoint: you predicted the overflow minutes before running the autoscale section.

WEEK 3 ☸️ Pods and Lambdas

Kubernetes and serverless scaling, then caching. Lab: change pod requests and count nodes; raise reserved concurrency; turn single-flight off and on.

🏁 Checkpoint: you can say how many pods, nodes and Lambda environments a load needs — and why the stampede hit the database.

WEEK 4 📚 Copies, splits and queues

Replicas and pools, partitioning, queues, surviving failure. Lab: shorten the replica lag; pick a better partition key; give the queue more workers; turn jitter and the circuit breaker off and on.

🏁 Checkpoint: 🏆 Capstone: scale the school results site for 10× traffic: CloudFront for the UI, a stateless API with an autoscaler, Redis in front of the database, replicas with a proxy, a queue for PDFs, timeouts, jittered retries and a breaker on every dependency — and write the load-test and failure-test plan that proves it.

🎓 13 / 13 — the fair survived results day!

Next: the Kubernetes school runs the counters, the Database school goes deeper on the register, and the School portal has the rest.