Tair (Enterprise Edition) DRAM-based instances support Bloom filters through the built-in Bloom module.
How Bloom filters work
A Bloom filter checks whether an element belongs to a dataset. The key tradeoff: a negative result (element absent) is definitive, but a positive result (element present) is probabilistic — a small fraction of positive answers will be false positives. This makes Bloom filters useful when you need fast exclusion checks and can tolerate a low false-positive rate.
Use cases
-
Web interception: Screen incoming requests against a known set, blocking those that clearly do not match any legitimate entry.
-
Cache penetration protection: Block requests for keys that don't exist in the database. A Bloom filter rejects clearly absent keys at the cache layer, preventing unnecessary load on downstream systems.
Supported instance type
Only Tair (Enterprise Edition) DRAM-based instances support Bloom filters.
Other built-in modules
DRAM-based instances ship with a set of self-developed Redis modules for complex data scenarios:
| Module | Description |
|---|---|
| exString | Extended string commands, including CAS and CAD operations |
| exHash | Extended hash commands |
| exZset | Extended sorted set commands |
| GIS | Geospatial indexing and queries |
| Bloom | Bloom filter for probabilistic membership checks |
| Doc | Document storage and retrieval |
| TS | Time series data |
| Cpc | Cardinality and frequency estimation |
| Roaring | Roaring Bitmap for efficient set operations |
| Search | Full-text and structured search |
| Vector | Vector similarity search |
These modules simplify application development in complex scenarios and allow you to focus on business innovation.