database-redis

Redis Best Practices

Priority: P0 (CRITICAL)

Redis is fast only when key shape, TTL ownership, and command complexity are explicit.

Rules

• Namespace keys by product/domain and entity.
• Give cache keys a TTL or a documented reason not to expire.
• Pick eviction policy to match workload: allkeys-lru for general caches, volatile-lru for mixed persistent/ephemeral data. Monitor hit rate and evictions.
• Avoid slow or unbounded commands in hot paths; prefer SCAN over KEYS and UNLINK over DEL for large-key deletion. lazyfree settings help background reclamation.

Verify

• Key naming identifies owner and invalidation scope.
• TTL or eviction policy exists for non-durable data.
• Large reads use bounded range/scan patterns.
• Connection timeouts and pool settings are explicit.
• Critical state does not exist only in Redis.

Anti-Patterns

• No sole truth in Redis: Always persist critical data to durable primary database.
• No large blobs: Split values > 100KB into smaller keys or use Hashes for field access.
• No JSON for objects: Use HSET for object fields to enable O(1) access without full decode.
• No TTL-less keys: Set TTL or eviction policy on all non-permanent keys to prevent unbounded growth.
• No KEYS in app paths: use SCAN or explicit index keys.

References

• Framework Map
• Best Practices Guide
• Checklist

Canonical response anchors

When this skill applies, preserve the following domain terminology or equivalent concrete examples in the answer when relevant:

• lazyfree