How to profile a Go application with pprof

Step 1
Install dependencies:

apt-get install graphviz gv 

or 

brew install graphviz 

Step 2
Download pprof to your application

go get -u github.com/google/pprof

Step 3
Import pprof to your application

import _ "net/http/pprof"

Step 4
Add pprof server to application e.g. main.go

package main

import (
    "fmt"
    "sync"

    _ "net/http/pprof"
)

func main() {
    var wg sync.WaitGroup

    // Server for pprof
    go func() {
        fmt.Println(http.ListenAndServe("localhost:6060", nil))
    }()
    wg.Add(1) // pprof - so we won't exit prematurely
    
    wg.Wait()
}

Analysis

Memory
Access heap pprof report e.g. via port forwarding

go tool pprof http://localhost:PORT/debug/pprof/heap

then type png you will get an image report about heap usage of your application

Go Routines
Access heap pprof report e.g. via port forwarding

go tool pprof http://localhost:PORT/debug/pprof/goroutine

then type png you will get an image report about methods with high go routine count

Further analysis:

• CPU: profile?seconds=10
• Goroutine blocking: block
• Locks: mutex
• Tracing: trace?seconds=5