go-pprof-example

Go Profiling Example

This repository contains a simple Go program demonstrating how to perform CPU and memory profiling using the runtime/pprof package.

Overview

The program performs the following actions:

1. Starts CPU profiling and writes the profile to a file named cpu.prof.
2. Runs a simple test function (testUser) that simulates some work by sleeping for 10 seconds.
3. Stops CPU profiling.
4. Optionally writes a memory profile to a file named mem.prof.

Prerequisites

• Go 1.16 or higher
• Basic knowledge of Go programming

Getting Started

Clone the Repository

git clone https://github.com/balasl342/go-pprof-examplee.git

Running the Program

To run the program, use the following command:

go run main.go

This will generate two files:

• cpu.prof: Contains the CPU profiling data.
• mem.prof: Contains the memory profiling data.

Analyzing the Profiles

You can analyze the generated profiles using the go tool pprof command.

CPU Profile

To analyze the CPU profile, run:

go tool pprof cpu.prof

This will open an interactive shell where you can explore the CPU usage data. Some useful commands within the pprof shell include:

• top: Shows the top functions by CPU usage.
• list <function>: Displays source code annotated with CPU usage for the specified function.
• web: Opens a graphical view of the profile in your web browser.

Memory Profile

To analyze the memory profile, run:

go tool pprof mem.prof

Similar to the CPU profile analysis, this will open the pprof shell for exploring memory usage data.

Cleaning Up

To remove the generated profiling files:

rm cpu.prof mem.prof

Code Structure

.
├── main.go        # Main Go file with profiling logic
└── README.md      # This README file

main.go

The main.go file contains the following key sections:

• CPU Profiling: Starts and stops CPU profiling, saving the data to cpu.prof.
• testUser Function: Simulates some work by sleeping for 10 seconds.
• Memory Profiling: Writes a heap profile to mem.prof after the CPU profiling is completed.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Go Documentation on pprof for detailed information on profiling in Go.
• Go tool pprof for profile analysis and visualization.