How to Install and Use k6 in Linux for Scriptable API and Web Performance Testing

Performance testing web applications, REST APIs, and microservices often presents a dilemma between cumbersome Java-based tools like Apache JMeter and overly simplistic command-line utilities like ab (ApacheBench) or wrk. Developed by Grafana Labs, k6 is a modern, developer-centric open-source load-testing framework written in Go that executes test scenarios scripted in standard ES6 JavaScript. It enables developers to simulate thousands of concurrent virtual users (VUs), inspect latencies with microsecond precision, and incorporate performance budgets into CI/CD pipelines.

Core Advantages of Grafana k6

Unlike legacy testing suites, k6 is built specifically for modern developer workflows:

• ES6 JavaScript Scripting: Write declarative test logic, dynamic payload generation, authentication headers, and assertions using modern JavaScript.
• High-Performance Go Engine: While tests are written in JavaScript, the underlying execution engine runs compiled Go goroutines, eliminating Node.js runtime overhead and supporting massive concurrency on minimal hardware.
• Thresholds and Performance Budgets: Define pass/fail SLA criteria (such as http_req_duration: ['p(95)<250']) that automatically fail your CI pipeline if latency exceeds limits.
• Structured Output Formats: Stream metric logs to standard terminal summaries, JSON files, Prometheus, InfluxDB, or Grafana dashboards.

Installing k6 on Linux Distributions

You can install k6 using official distribution repositories, standalone binaries, or Docker:

Method 1: Install on Ubuntu and Debian

# Import official Grafana GPG key
sudo gpg -k
sudo gpg --no-default-keyring --keyring /usr/share/keyrings/k6-archive-keyring.gpg \
  --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys C5AD17C747E3415A3642D57D77C6C491D34EE24C

# Add repository
echo "deb [signed-by=/usr/share/keyrings/k6-archive-keyring.gpg] https://dl.k6.io/deb stable main" \
  | sudo tee /etc/apt/sources.list.d/k6.list

# Update and install
sudo apt update && sudo apt install k6

Method 2: Install on Fedora, RHEL, and AlmaLinux

sudo dnf install https://dl.k6.io/rpm/repo.rpm
sudo dnf install k6

Method 3: Install on Arch Linux

sudo pacman -S k6

Writing Your First k6 Load Test Script

Create a file named load_test.js and define a baseline testing scenario:

import http from 'k6/http';
import { check, sleep } from 'k6';

// Define execution stages and performance thresholds
export const options = {
  stages: [
    { duration: '30s', target: 50 },  // Ramp-up to 50 virtual users
    { duration: '1m', target: 50 },   // Stay at 50 users
    { duration: '30s', target: 0 },   // Ramp-down to 0 users
  ],
  thresholds: {
    http_req_duration: ['p(95)<300'], // 95% of requests must complete under 300ms
    http_req_failed: ['rate<0.01'],    // Error rate must remain below 1%
  },
};

export default function () {
  const url = 'https://httpbin.test.k6.io/get';
  const response = http.get(url);

  // Validate response integrity
  check(response, {
    'status is 200': (r) => r.status === 200,
    'protocol is HTTP/2': (r) => r.proto === 'HTTP/2.0',
  });

  sleep(1);
}

Running Tests and Interpreting Terminal Metrics

Execute the test script directly from your terminal:

k6 run load_test.js

Upon completion, k6 outputs a comprehensive terminal summary breakdown:

Metric Name	Type	Significance
http_req_duration	Trend	Total time taken for request round-trips (avg, min, med, p90, p95).
http_req_waiting	Trend	Time spent waiting for the first byte of response (TTFB).
http_req_failed	Rate	Percentage of requests that returned HTTP 4xx or 5xx error codes.
iterations	Counter	Total complete executions of the default function across all VUs.
vus_max	Gauge	Peak allocated virtual user concurrency during the test run.

Exporting Results to JSON for CI Automation

To archive performance data or pass metrics to reporting tools:

k6 run --summary-export=summary.json --out json=detailed_metrics.json load_test.js