Backpressure
A flow-control condition in which a downstream consumer cannot process messages as fast as an upstream producer publishes them, causing queue depth to grow. Detected as the BACKPRESSURE_ACCUMULATION operational primitive. Unaddressed backpressure leads to lag growth, latency expansion, and eventually broker or consumer saturation.
Backpressure is one of the most common and consequential operational primitives the Cognitive Substrate monitors. It appears when a Kafka consumer group falls behind its topic's write rate - either because processing logic has become more expensive, resource contention has reduced throughput, or upstream volume has spiked. The pattern library treats BACKPRESSURE_ACCUMULATION as a precursor to several downstream failure patterns, including TAIL_LATENCY_EXPANSION and QUEUE_GROWTH. Early detection through consumer lag monitoring and system mapping allows the recommendation engine to suggest interventions (scaling replicas, increasing partition count, throttling producers) before the condition becomes critical. In KEDA-managed deployments, consumer lag is the primary signal that drives autoscaling of cognitive workers.