CPU profiling basics in Node.js - Step-by-Step Execution

Concept Flow - CPU profiling basics

Start Node.js app

↓

Enable CPU profiler

↓

Run workload

↓

Collect CPU profile data

↓

Analyze profile

↓

Identify slow functions

↓

Optimize code

↓

Repeat profiling if needed

This flow shows how CPU profiling works: start the app, enable profiling, run code, collect data, analyze to find slow parts, then optimize.

Execution Sample

Node.js

import * as inspector from 'node:inspector';

const session = new inspector.Session();

inspector.open(9229);

await session.connect();

await session.post('Profiler.enable');
await session.post('Profiler.start');

// run code

const { profile } = await session.post('Profiler.stop');

console.log(profile);

This code starts the CPU profiler, runs some code, stops profiling, and logs the collected CPU profile data.

Execution Table

Step	Action	Profiler State	CPU Usage Data	Output
1	Start Node.js app	Profiler inactive	No data	App running
2	Enable CPU profiler	Profiler active	Collecting data	Profiler started
3	Run workload	Profiler active	Recording CPU samples	Workload executing
4	Stop CPU profiler	Profiler inactive	Final CPU profile collected	Profile data ready
5	Analyze profile	Profiler inactive	Profile data analyzed	Identified slow functions
6	Optimize code	Profiler inactive	No data change	Code improved
7	Repeat profiling if needed	Profiler active	New data collected	Cycle continues

💡 Profiling stops when profiler is disabled or app ends

Variable Tracker

Variable	Start	After Step 2	After Step 4	After Step 5	Final
profilerState	inactive	active	inactive	inactive	inactive
cpuProfileData	none	collecting	collected	analyzed	analyzed

Key Moments - 3 Insights

Why does the profilerState change from active to inactive after stopping?

Because stopping the profiler disables data collection, so profilerState changes to inactive as shown in execution_table step 4.

What does cpuProfileData contain after stopping the profiler?

It contains the full CPU usage profile collected during the workload, ready for analysis as shown in execution_table step 4.

Why do we need to repeat profiling after optimizing code?

To verify if the optimizations improved performance by collecting new CPU data, as shown in execution_table step 7.

Visual Quiz - 3 Questions

Look at the execution_table, what is the profilerState at Step 3?

A inactive

B active

C starting

D stopped

💡 Hint

Check the 'Profiler State' column at Step 3 in the execution_table.

At which step does the CPU profile data become fully collected?

A Step 2

B Step 3

C Step 4

D Step 5

💡 Hint

Look at the 'CPU Usage Data' column to find when data collection finishes.

If we skip Step 6 (Optimize code), what happens at Step 7?

A New profile data will still be collected

B Profiler will remain inactive

C No new data will be collected

D App will crash

💡 Hint

Step 7 shows profiler active and new data collected regardless of optimization.

Concept Snapshot

CPU Profiling Basics in Node.js:
- Start profiler before workload
- Run code to collect CPU samples
- Stop profiler to get profile data
- Analyze to find slow functions
- Optimize code and repeat profiling
Use node:inspector Profiler API for control.

Full Transcript

CPU profiling in Node.js involves starting the profiler before running your code, collecting CPU usage data during execution, then stopping the profiler to get the full profile. This profile shows which functions use the most CPU time. You analyze this data to find slow parts of your code. After optimizing those parts, you can profile again to check improvements. The profiler state changes from inactive to active when started, and back to inactive when stopped. CPU profile data is collected only while the profiler is active. This process helps improve app performance by focusing on real CPU usage.