Load Testing with Multiple Cloud Region Traffic Simulation

Load test from several regions at once

One load generator in one data centre measures your server plus one network path. Real traffic comes from many places, and the slow requests are usually the ones that cross an ocean, miss the CDN, or hit a region your auto-scaling forgot. This template runs the same test from several LoadFocus cloud regions at the same time and shows response time and error rate per region, so you can see whether a slow p95 is your application or your geography.

Configuration

Run this templateOpens the cloud test form with these values filled in. Free plan runs it at the free user limit; sign in or create a free account first.

The button prefills virtual users, duration and ramp-up on the cloud test form. Add your URLs, then pick the regions in the location selector (multi-region runs need a paid plan; the free plan runs from one region).

What to read in the results

• Response time by location. The Overview tab lets you switch between combined and per-location numbers. A region that is 300 ms slower at 10 users and 2 s slower at 400 is a network or CDN problem; one that matches the others until the ramp finishes and then diverges is a regional capacity problem.
• Error rate by status code. 5xx that only appears in one region points at a regional dependency (a single-region database, a rate limiter in front of one edge). 429s everywhere mean your CDN or WAF is throttling the test, not your origin.
• Throughput plateau. When adding users stops adding requests per second, you have found the ceiling for that region.

Pass/fail thresholds for this template

Set these on the test once and every run gets a PASS or FAIL verdict; the same verdict is available to CI through the API and the GitHub Action.

The same scenario as a k6 script

If you prefer scripts, upload this to a k6 cloud test. Regions are chosen in the LoadFocus UI, not in the script, so the script only describes the load shape.

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

export const options = {
  stages: [
    { duration: '2m', target: 500 },  // ramp-up
    { duration: '8m', target: 500 },  // hold
  ],
  thresholds: {
    http_req_duration: ['p(95)<800'],
    http_req_failed: ['rate<0.01'],
  },
};

const pages = ['/', '/search?q=shoes', '/api/products?limit=20', '/static/app.css'];

export default function () {
  const url = 'https://www.example.com' + pages[Math.floor(Math.random() * pages.length)];
  const res = http.get(url);
  check(res, { 'status is 200': (r) => r.status === 200 });
  sleep(1 + Math.random() * 2);
}

When to run it

• Before a launch or campaign that will bring traffic from a new market.
• After adding or moving a CDN, a region, or a database replica.
• On a schedule (weekly is enough) so a regional regression shows up as a red run in the Trend tab, not in support tickets.

FAQ on multi-region load testing

How many regions should I use?

Start with three: your two biggest markets and one you do not serve, as a control. Add more once the first run shows a clear difference between regions.

Does the free plan run multi-region tests?

The free plan runs from one region at a reduced user count. Multi-region runs and the 500-user configuration above need a paid plan.

Why is one region slower even at low load?

That is latency and routing, not capacity: distance to your origin, a CDN cache miss, or DNS pointing that region at a far-away endpoint. It shows up at 1 user and stays roughly constant as load grows.

Can I weight the traffic per region?

Yes. On paid plans the location selector lets you split the virtual users by percentage, for example 60% US, 30% EU, 10% APAC, to mirror your real audience.