Related: https://en.wikipedia.org/wiki/Log-structured_merge-tree https://en.wikipedia.org/wiki/Log-structured_merge-tree >"LSM trees are used in data stores s

Related: https://en.wikipedia.org/wiki/Log-structured_merge-tree https://en.wikipedia.org/wiki/Log-structured_merge-tree >"LSM trees are used in data stores such as Apache AsterixDB, Bigtable, HBase, LevelDB, Apache Accumulo, SQLite4,[8] Tarantool,[9] RocksDB, WiredTiger,[10] Apache Cassandra, InfluxDB,[11] YugabyteDB,[12] ScyllaDB, and CockroachDB." Basically, these are databases that retain performance for write-heavy workloads (which should include most NoSQL database implementations), i.e., database write requests far exceed database read requests... LSM Trees appear to be the principal database data structure for implementing such write-heavy performance... Related: Starting from Zero: Build an LSM Database with 500 Lines of Code (2021): https://www.alibabacloud.com/blog/starting-from-zero-build-an-lsm-database-with-500-lines-of-code_598114 https://www.alibabacloud.com/blog/starting-from-zero-build-a...