Layer 7 is the named arguments and laws that show up in design reviews: end-to-end, Conway, strangler fig, tail latency, and gossip.

Each node tells a few peers what it knows. Eventually the cluster agrees, without a single coordinator.

If one request fans out to 50 backends, the slowest of those 50 is your user-visible latency.

Do not rewrite the monolith in one shot. Intercept calls, replace slice by slice, then retire the core.

You ship your org chart. Service boundaries that fight how people talk will keep rotting.

The network or middleware cannot fully guarantee the application is correct. The ends still have to check.

An SLI is the measurement, an SLO is the target, and the error budget is how much unreliability you will spend on change.

Shard the whole system into independent cells so a bad deploy or hot shard takes down a slice, not the planet.

Run a companion container (proxy, agent) beside the app so mesh, secrets, and telemetry are not baked into the binary.

Data structures designed so replicas can update independently and always merge to the same result.

A compact structure that can say 'definitely not present' or 'maybe present' — never a false negative, sometimes a false positive.

Two separate writes cannot both succeed or both fail together. One will commit and the other will not.

Put the 'tell the rest of the world' record in an outbox table in the same commit as the source-of-truth write, then publish asynchronously.

You cannot trust NTP to order events. Logical clocks track causality without a shared timeline.

BASE is the slogan for stores that stay up by allowing replicas to disagree for a while.

A generation number or timed lease lets storage reject writes from a leader that does not know it was replaced.

Paxos and Raft are the usual answers to 'how do we pick one leader and one history?' in a crash-stop cluster.

A network cut can leave two leaders, two writers, and diverging data until you fence one side.

CAP only talks about partitions. PACELC adds the everyday tradeoff: even when the network is fine, you still pick latency or stronger consistency.

In an asynchronous network, crash-stop consensus cannot be both always safe and always live. Timeouts are how we cheat.