Open this lesson in your favourite AI. It'll walk you through the why, explain the demo, and quiz you on the try-it list.
Latency, redundancy, data sovereignty, and price all change across regions — and within a region, AZs are your unit of fault isolation. Pick the wrong region and you ship a 250ms round-trip to your users; pick a single AZ and one transformer fire takes you offline. Edge locations and CDN POPs are a separate fabric layered on top, optimized for read-heavy traffic.
Enumerate regions and zones on each cloud and measure round-trip latency from your laptop to a couple of them.
Use these three in order. Each builds on the one before.
In one paragraph, explain regions vs availability zones vs edge locations like I'm new to cloud, with a concrete example of when each matters.
Walk me through how a multi-AZ RDS failover actually works step by step — what triggers it, how DNS gets repointed, and what the application sees during the cutover.
Given a global SaaS with users in India, Germany, and Brazil that needs <100ms p95 reads and a single source of truth for writes, design the region + edge layout and justify each placement.
# All regions, then AZs in one
aws ec2 describe-regions --query 'Regions[*].RegionName' --output text | tr '\t' '\n' | head -10
aws ec2 describe-availability-zones --region ap-south-1 \
--query 'AvailabilityZones[*].{name:ZoneName,id:ZoneId,state:State}' \
--output table
# Latency probe — DNS-only, no auth needed
for r in us-east-1 eu-west-1 ap-south-1 ap-southeast-1; do
echo -n "$r: "
curl -s -o /dev/null -w "%{time_connect}s\n" https://ec2.$r.amazonaws.com/
done