README

Continuous Profiling for Node applications

Profiling a Node Rideshare App with Pyroscope

Note: For documentation on Pyroscope's Node integration visit our website.

Background

In this example, we show a simplified, basic use case of Pyroscope. We simulate a "ride share" company which has three endpoints found in main.js:

• /bike : calls the bikeSearchHandler() function to order a bike
• /car : calls the carSearchHandler() function to order a car
• /scooter : calls the scooterSearchHandler() function to order a scooter

We also simulate running 3 distinct servers in 3 different regions (via docker-compose.yml)

• us-east
• eu-north
• ap-south

One of the most useful capabilities of Pyroscope is the ability to tag your data in a way that is meaningful to you. In this case, we have two natural divisions, and so we "tag" our data to represent those:

• region: statically tags the region of the server running the code

Tagging static region

Tagging something static, like the region, can be done in the initialization code in the main() function:

  Pyroscope.init({
    appName: 'nodejs',
    serverAddress: process.env['PYROSCOPE_SERVER'] || 'http://pyroscope:4040',
    tags: { region: process.env['REGION'] || 'default' }
  });

Resulting flame graph / performance results from the example

Running the example

There are 3 examples:

• express - basic integration example
• express-ts - type script example
• express-pull — pull mode example

To use any of the examples, run the following commands:

# change directory
cd express # or cd express-ts / cs express-pull

# Pull latest pyroscope and grafana images:
docker pull grafana/pyroscope:latest
docker pull grafana/grafana:latest

# Run the example project:
docker-compose up --build

# Reset the database (if needed):
docker-compose down

This example runs all the code mentioned above and also sends some mock-load to the 3 servers as well as their respective 3 endpoints. If you select nodejs.wall from the dropdown, you should see a flame graph that looks like this (below). After 20-30 seconds, the flame graph updates. Click the refresh button. We see our 3 functions at the bottom of the flame graph taking CPU resources proportional to the size of their respective search_radius parameters.