Profiling Go applications with pprof

Identifying CPU hotspots and memory allocation sources in a running Go application requires instrumentation. Guessing is ineffective.

Enable the pprof HTTP endpoint and use the go tool:

import (
    "net/http"
    _ "net/http/pprof" // registers /debug/pprof endpoints
)

func main() {
    // In a goroutine alongside your main server
    go http.ListenAndServe(":6060", nil)
    // ... rest of app
}

# CPU profile: sample for 30 seconds
go tool pprof http://localhost:6060/debug/pprof/profile?seconds=30

# Memory (heap) profile
go tool pprof http://localhost:6060/debug/pprof/heap

# Interactive web UI (requires graphviz)
go tool pprof -http=:8080 cpu.prof

pprof samples the call stack at regular intervals (CPU) or records heap allocations. The resulting profile shows where time and memory are spent, enabling data-driven optimisation.