Understanding Redis Big Keys: Definition, Risks, Detection, and Deletion Strategies

What is a Redis big key?

A big key refers to a key whose value size is large, not the length of the key name. Typical thresholds are about 15 KB for a String value and roughly 1,500 elements for collection types such as zset, list, hash, or set.

Why big keys are dangerous

When a key grows too large it can consume excessive network bandwidth, cause request timeouts that block subsequent operations, and increase memory pressure leading to out‑of‑memory errors. In a Redis cluster, deleting a big key may take a long time, potentially triggering master‑slave failover or breaking replication.

Impact on the cluster

Long deletion times stall the master node, which can cause a failover or interrupt synchronization between nodes.

How to identify big keys

Third‑party tools can parse dump.rdb snapshot files to locate big keys. The following command finds keys larger than 20 KB: dump.rdb -c memory --bytes 20480 -f redis.csv Redis also provides a built‑in command: redis-cli --bigkeys This command lists big‑key information directly.

How to delete big keys safely

Use ITRIM (or LTRIM for lists) to remove a small number of elements at a time.

Use SSCAN to scan about 150 set members, then delete each with SREM.

Use ZREMRANGEBYRANK to remove 150 elements from a sorted set in a single call.

Use HSCAN to fetch roughly 150 hash fields, then delete each with HDEL.

Redis 4.0+ supports asynchronous deletion; replace DEL with UNLINK so the key is removed in the background without blocking the main thread.

These incremental approaches avoid blocking the server while removing large values.