liveBook · Manning

Backpressure

As explained in Build High Value Software Systems

A mechanism by which a downstream system signals that upstream work should be slowed or limited to match its processing capacity.

When a producer sends work faster than a downstream service can handle it, the service may fail or drop requests. Backpressure lets the consumer set the pace: it signals the producer to slow down when capacity is limited. section 11.5.1 section 11.5

A bounded queue provides a simple mechanism. Consumers pull work at a sustainable rate; when the queue is full, publishing blocks until capacity becomes available. Queue pressure, expressed from 0.0 to 1.0, makes capacity visible: high pressure can prompt alerts, dashboards, scaling, or producer throttling. section 11.5.1

Reading queue pressure

A queue at 0.1 pressure has headroom; at 0.8, the consumer is falling behind; at 1.0, the producer waits. A synchronous chain can still run internally while a queue controls the pace of work entering it. section 11.5.1

How Build High Value Software Systems covers it

How deep it goes
Defines bounded-queue blocking, pressure levels, monitoring, and scaling responses, with a small Python implementation.
Examples
Uses a bounded queue with pressure readings of 0.1, 0.8, and 1.0, plus a Python asyncio implementation.
What you can do after
Apply via a bounded queue to block publishing at capacity and route pressure metrics to alerts, dashboards, or scaling triggers.
How it connects
Connects pull architectures, message queues, producer-consumer pacing, and bottleneck visibility.

Also explained in

Compare how 10 books explain backpressure →

Related concepts

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