What's a load balancer?
Software that spreads incoming visitors across several servers so no single one gets overwhelmed. If a server dies, it just stops sending traffic there.
What does "auto-scaling" mean?
The system adds or removes servers on its own based on how much traffic shows up — more visitors, more servers; quiet periods, fewer (and cheaper).
What is Kubernetes, really?
A system that runs and organizes many small app copies ("pods") across a cluster of machines, and automatically restarts anything that crashes — without a person doing it by hand.
What's the difference between a database and a cache?
The database is where data actually lives, permanently. The cache is a fast, temporary copy of the most-requested data, so the database doesn't get hit on every single request.
Why run in multiple regions or clouds?
If one data center — or an entire cloud provider — goes down, traffic automatically reroutes to a healthy one. The app stays up even through a major outage.
What does "self-healing" actually mean?
The system notices on its own when something has crashed and replaces it — no one has to get paged and manually restart a server at 3am.
Why start with a monolith at all?
It's simple and cheap to run. The tradeoff is that everything shares one failure domain — if that one server goes down, the entire app goes down with it.
Are the cost and revenue numbers real?
They're illustrative, not a quote — the point is to show scaling and redundancy as real dollar tradeoffs, not just boxes and arrows.