Databases: How do they Work

A. Redis

1. Primarily single-threaded in-memory datastore. Commands are executed sequentially, eliminating the need for locks and enabling very low latency.
2. Key-value datastore supporting rich data structures such as strings, lists, sets, hashes, and sorted sets.
3. Persistence is optional. Redis supports:
   • RDB snapshotting
   • Append Only File (AOF) logging (fsync -> 1 second)
4. Scaling strategies:
   • Replication (master for writes → replicas for read scaling)
   • Sharding using Redis Cluster for horizontal scaling.
5. Cache eviction policies include:
   • LRU (Least Recently Used)
   • LFU (Least Frequently Used)
   • TTL-based eviction

B. Elastic-Search (Search-Engine)

C. InfluxDB

D. DyanamoDb

E. CassandraDb

F. MongoDB

G. Sql