Consistent Hashing

Summary: A way of evenly distributing load across an internet-wide system of caches or nodes while minimizing data movement when nodes are added or removed.

Sources: chapter6

Last updated: 2026-04-15


Consistent hashing (defined by Karger et al.) uses randomly chosen partition boundaries to avoid the need for central control or distributed consensus. (source: chapter6)

Note on Terminology

In the context of databases, the term “consistent hashing” is often used loosely to refer to any hash-partitioning strategy, but it has a specific meaning in academic literature. Kleppmann notes that it doesn’t work very well for databases in its original form and is rarely used in practice; instead, “hash partitioning” is a more accurate term for most database implementations. (source: chapter6)