Caching, Redis & Cache Failure Modes
Question bank: caching trade-offs, Redis internals, Redis Cluster vs Sentinel failover mechanics, stampede, penetration, avalanche, hot keys, warming and stale-while-revalidate — fresh angles for 2026 loops.
About 120 minutes · 24 exam questions · 23 interview questions · 2 code labs
What you will learn
- Caching — when it pays, and when it makes things worse
- Redis — single-threaded execution, TTL semantics and durability limits
- Redis Cluster — slots, redirects, hash tags and failover
- Redis Sentinel — HA for one primary, not sharding
- Cache stampede — what the waiters do matters as much as the lock
- Cache penetration — misses the cache can never absorb
- Cache avalanche — many keys, or the whole tier, at once
- Hot keys — detect first, then pick a read or a write fix
- Cache warming — pre-fill deliberately, without becoming the incident
- Stale-while-revalidate — soft TTL, hard TTL, one refresher
Previous: Load Balancing, Gateways, Rate Limits, CDN & Multi-tenancy · Next: Indexes, Pools, Replication, Sharding, Transactions & Migrations