How to run production systems reliably, as Google's SRE books and common industry practice describe it: reliability targets and error budgets, alerting, incident response, postmortems, capacity, resilience patterns and chaos testing.
You’re reading a free preview. Sign up to see every topic and question — and to play this deck as games.
How to choose what to measure, set a target for it, write a contract around it, and work out how much downtime each level of availability allows.
39 total questions
Card #1
Service level indicator (SLI)
Card #2
Service level objective (SLO)
Card #3
Service level agreement (SLA)
1. A pipeline publishes a sales report every hour, and users complain that the numbers are often out of date. Which SLI fits their complaint best?
2. Your availability SLO is being met, yet users complain that checkout feels slow. What is missing from your SLOs?
3. A service fails every request for 10 minutes at 3 am, when traffic is a tenth of normal. Which kind of SLI shows this outage as more serious?
1. A common SLI is the number of good events divided by the number of ___ events.
Hint: not every request is counted
2. When a provider misses an SLA, customers are usually given service ___.
Hint: money off a future bill
3. The 50th percentile latency is also called the ___.
Hint: the middle value
