DEV Community

Cover image for How to send more requests with variable payload size in K6?
CodeWithVed
CodeWithVed

Posted on

How to send more requests with variable payload size in K6?

#k6 #autoscaling #performance #testing

This k6 script is designed for load testing an application by simulating multiple virtual users (VUs) that perform a series of HTTP requests. It includes setup for authentication, payload generation, and defines different testing scenarios.

Script Breakdown

  • Imports and Setup: The script imports the http module for making HTTP requests and the check function for validating responses. It also imports a service_name function from an external module. The setup function is executed once before the main test runs. It sends a POST request to a token service to obtain an access token, checking that the response status is 200 (OK) and returning the token for use in subsequent requests. 
export function setup() { const url = 'https://example-token-service.com/token'; const formData = {}; const loginRes = http.post(url, formData); check(loginRes, { 'response code was 200': (res) => res.status === 200, }); return { token: loginRes.json().access_token }; }
Enter fullscreen mode Exit fullscreen mode
  • Payload Generation: The script defines a target size for the payload (700 KB) and generates a JSON payload array using the generatePayloadArray function. This function constructs an array of empty objects until the total size reaches the specified target. 
const targetSizeKB = 700; const payload = generatePayloadArray(targetSizeKB);
Enter fullscreen mode Exit fullscreen mode
  • Scenarios Definition: Two scenarios are defined: 
perVuIterations: 20 VUs, each performing 5000 iterations over a maximum duration of 300 seconds. constantRequestRate: A constant rate of 248 requests per second for 300 seconds, with a maximum of 20 VUs. const scenarios = { perVuIterations: { executor: 'per-vu-iterations', vus: 20, iterations: 5000, maxDuration: '300s', }, constantRequestRate: { executor: 'constant-arrival-rate', rate: 248, timeUnit: '1s', duration: '300s', preAllocatedVUs: 20, maxVUs: 20, }, };
Enter fullscreen mode Exit fullscreen mode
  • Options Configuration: The script configures options for the test, including skipping TLS verification and selecting scenarios based on an environment variable. 
export const options = { insecureSkipTLSVerify: true, scenarios: SCENARIO ? { [SCENARIO]: scenarios[SCENARIO] } : {}, };
Enter fullscreen mode Exit fullscreen mode
  • Main Function: The default function is executed by each VU, sending a POST request to the specified application endpoint using the token and payload generated earlier. 
export default function (data) { const url = 'https://your-application-endpoint.com/_POST'; service_name(url, data.token, payload); }
Enter fullscreen mode Exit fullscreen mode
  • Payload Generation Function: The generatePayloadArray function constructs a JSON object containing an array of payloads, ensuring the total size matches the target size. 
function generatePayloadArray(targetSizeKB) { const payloadArray = []; let currentSize = 0; while (currentSize < targetSizeKB * 1024) { const payload = {}; const jsonString = JSON.stringify(payload); payloadArray.push(payload); currentSize += jsonString.length; } return JSON.stringify({ payload: payloadArray }); }
Enter fullscreen mode Exit fullscreen mode

This script effectively sets up a load test that simulates user interactions with an application, allowing for performance evaluation under various conditions.
Complete Script as Follow:

 import http from 'k6/http'; import { service_name } from './service.js'; import { check } from 'k6'; export function setup() { const url = 'https://example-token-service.com/token'; const formData = {}; const loginRes = http.post(url, formData); check(loginRes, { 'response code was 200': (res) => res.status === 200, }); return { token: loginRes.json().access_token }; } const targetSizeKB = 700; const payload = generatePayloadArray(targetSizeKB); const scenarios = { perVuIterations: { executor: 'per-vu-iterations', vus: 20, iterations: 5000, maxDuration: '300s', }, constantRequestRate: { executor: 'constant-arrival-rate', rate: 248, timeUnit: '1s', // 248 iterations per second, i.e. 248 RPS duration: '300s', preAllocatedVUs: 20, // initial pool of VUs maxVUs: 20, // max VUs if preAllocatedVUs are not enough }, }; const { SCENARIO } = __ENV; export const options = { insecureSkipTLSVerify: true, scenarios: SCENARIO ? { [SCENARIO]: scenarios[SCENARIO] } : {}, }; export default function (data) { const url = 'https://your-application-endpoint.com/_POST'; service_name(url, data.token, payload); } function generatePayloadArray(targetSizeKB) { const payloadArray = []; let currentSize = 0; while (currentSize < targetSizeKB * 1024) { const payload = {}; const jsonString = JSON.stringify(payload); payloadArray.push(payload); currentSize += jsonString.length; } return JSON.stringify({ payload: payloadArray }); }
Enter fullscreen mode Exit fullscreen mode
  • If you want to gradually increase the load, you can utilize the provided script, which implements a ramp-up strategy.. 
 import http from 'k6/http'; import { check } from 'k6'; export let options = { stages: [ { duration: '1m', target: 80 }, { duration: '1m', target: 100 }, ], insecureSkipTLSVerify: true, }; export function setup() { const url = 'https://get-token-service.com/token'; const res = http.post(url, {}); return { token: res.json().access_token }; } export default function ({ token }) { const url = 'https://example-server.com/_POST/api/endpoint'; const headers = { 'Authorization': token, 'Content-Type': 'application/json', }; const body = JSON.stringify({ payload: [] }); const res = http.post(url, body, { headers }); check(res, { 'status was 202': (r) => r.status === 202 }); }
Enter fullscreen mode Exit fullscreen mode

Top comments (0)