AWS, Azure & GCP · Cloud across providers
How do you monitor and log a cloud system?
Short answer
Collect metrics, logs and traces into a central platform, define SLOs and alert on symptoms, and use dashboards for diagnosis. Cloud-native tools include CloudWatch, Azure Monitor and Cloud Operations.
Metrics tell you that something is wrong, logs tell you what happened, and traces tell you where time went across services. A useful observability setup correlates all three with shared identifiers such as request id and trace id.
What to instrument
- Golden signals: latency, traffic, errors and saturation
- SLOs with error budgets, and alerts that fire on user-facing symptoms
- Structured logs with consistent fields, not free-text messages
- Distributed traces across synchronous calls and asynchronous consumers
- Dashboards for diagnosis, not as the primary alerting mechanism
logger.info("order_created", {
orderId: order.id,
userId: order.userId,
durationMs: Date.now() - startedAt,
traceId: context.traceId,
});Alert fatigue is an architectural problem. If every CPU spike pages someone, real incidents get missed. Alert on SLO burn rate or sustained error and latency symptoms, and route low-urgency signals to a dashboard or ticket.
Follow-ups you should expect
Interviewers use these to check whether the first answer was understanding or memorisation.
- 1What are the four golden signals?
- 2Why alert on SLO burn rate rather than raw CPU?
- 3How do you correlate logs across services?
- 4How do you control observability cost?
Say it in the interview
I collect metrics, logs and traces centrally, correlate them with request and trace ids, and alert on user-facing SLO symptoms rather than every infrastructure spike. Dashboards are for diagnosis, not the primary pager. I also redact sensitive fields at the source and set retention deliberately, because observability data gets expensive and risky if left unchecked.
Use this as a model, not a script. Saying it in your own words is what makes it survive follow-up pressure.
Related questions
Reading is step one
Now say it out loud.
Understanding an answer and delivering it under pressure are different skills. Rehearse this topic in a live AWS, Azure & GCP mock interview and get scored feedback on depth, clarity and confidence.