Scaling Goroutines & Concurrency Patterns in Production

Deep dive into goroutine lifecycle management, channel patterns, worker pools, and memory leaks in high-load Go microservices.

GopherCon Europe 2026📍 Berlin, Germany
goconcurrencyperformancesystems

Talk Overview

Concurrency in Go is cheap, but goroutines are not free. In high-throughput distributed systems, unmanaged goroutine spawning leads to runaway memory usage, GC pressure, and latent goroutine leaks.

Topics Covered

  1. Channel ownership patterns: who closes what and when.
  2. Context propagation, cancellation deadlines, and cascading timeouts.
  3. Bounded worker pools vs dynamic fan-out workers.
  4. Diagnosing production leaks using net/http/pprof and goroutine stack dumps.