Find agent skills by outcome

### Redis (Clusters

Sentinel

- Set TTLs based on data change frequency: seconds for real-time data

minutes for session data

- Enable compression (LZ4

Snappy) for large values where CPU overhead is acceptable

- Choose binary serialization (Protocol Buffers

MessagePack) over JSON for reduced size and faster parsing

- Use hierarchical namespaced keys (e.g.

`app:user:123:profile`) for logical grouping and bulk invalidation

- Avoid embedding volatile data (timestamps

random values) in keys that should be shared

- [ ] Security controls (TLS

authentication

- [ ] Memory sizing accounts for peak load

data growth

- [ ] Monitoring covers hit rates

latency

- [ ] Cache hit rates meet target thresholds (typically >90% for hot data

>70% for warm data)

Creative

- Design load tests simulating cache stampede

hot key

- Create benchmark suite measuring cache hit rates

latency percentiles (p50

Business

After designing or modifying a caching strategy

verify:

- Define alerting thresholds for cache health (hit rate drops

latency spikes

Writing & Content

- Map each data entity to its invalidation strategy (write-through

write-behind

- Document cache key schema with namespaces

versioning

- Cost optimization: Right-sizing cache instances

tiered storage (hot/warm/cold)

- High availability: Redis Sentinel or Cluster mode

Memcached replication

- Security: Encryption in transit (TLS)

authentication (Redis AUTH

- Cache stampede prevention: Mutex locks

probabilistic early expiration

Business

- Memory management: Eviction policy selection

maxmemory configuration

- Hot key mitigation: Local caching of hot keys

key replication across shards

Writing & Content

- Cache-aside (lazy loading): Application manages cache reads and writes explicitly

simple but risk of stale reads

Writing & Content

- Write-behind (write-back): Asynchronous batch writes to backing store

lower write latency