liveBook · Manning

Soft state

As explained in Grokking Software Architecture

Soft state is distributed system state that can change over time without new input as replicas synchronize.

Soft state is distributed data that may change over time without new input as replicas synchronize. In the NoSQL BASE tradeoff, it accompanies availability and scale, while readers may see stale or out-of-sync values. section 8.2.2

When stale data is acceptable

A social-media like count may tolerate soft state; financial balances and inventory are risky places to rely on it because readers may see stale or out-of-sync data. section 8.2.2

How Grokking Software Architecture covers it

How deep it goes
Names soft state within BASE and notes stale reads, but does not explain synchronization mechanics.
Examples
Contrasts a social-media like count with financial balances and inventory.
What you can do after
Evaluate whether stale or out-of-sync reads are acceptable for a use case.
How it connects
Connects soft state to NoSQL, BASE, eventual consistency, availability, horizontal scaling, and ACID.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Software Architecture. The section links point to the parts of the book it draws on.