Skip to main content

Gatling Performance Testing

Gatling is a powerful performance testing tool designed for high-performance load testing of web applications and APIs.

🎯 Why Gatling?​

  • High Performance - Can simulate thousands of users
  • Scala-based - But easy to use without Scala knowledge
  • Real-time Monitoring - Live metrics during test execution
  • Beautiful Reports - Rich HTML reports with charts
  • CI/CD Integration - Easy integration with build pipelines

πŸ“¦ Installation​

Option 1: Standalone Bundle​

Download from Gatling Downloads

Option 2: Maven Integration​

Add to your pom.xml:

<plugin>
    <groupId>io.gatling</groupId>
    <artifactId>gatling-maven-plugin</artifactId>
    <version>4.3.7</version>
</plugin>

πŸš€ Your First Gatling Test​

Create BasicSimulation.scala:

import io.gatling.core.Predef._
import io.gatling.http.Predef._
import scala.concurrent.duration._

class BasicSimulation extends Simulation {

  val httpProtocol = http
    .baseUrl("https://jsonplaceholder.typicode.com")
    .acceptHeader("application/json")

  val scn = scenario("Basic API Test")
    .exec(
      http("Get Users")
        .get("/users")
        .check(status.is(200))
    )
    .pause(1)
    .exec(
      http("Get Posts")
        .get("/posts")
        .check(status.is(200))
        .check(jsonPath("$[0].title").exists)
    )

  setUp(
    scn.inject(
      atOnceUsers(10),           // 10 users at once
      rampUsers(50) during (30.seconds)  // 50 users over 30 seconds
    )
  ).protocols(httpProtocol)
}

πŸ”§ Advanced Features​

Load Patterns​

setUp(
  scn.inject(
    nothingFor(4.seconds),                    // Wait 4 seconds
    atOnceUsers(10),                          // 10 users immediately
    rampUsers(50) during (30.seconds),        // Ramp up 50 users over 30s
    constantUsersPerSec(20) during (60.seconds), // Constant rate
    heavisideUsers(1000) during (20.seconds)  // Gradual ramp
  )
)

Data Feeders​

val userFeeder = csv("users.csv").random

val scn = scenario("User Creation")
  .feed(userFeeder)
  .exec(
    http("Create User")
      .post("/users")
      .body(StringBody("""{"name": "${name}", "email": "${email}"}"""))
      .header("Content-Type", "application/json")
  )

Assertions​

setUp(scn.inject(rampUsers(100) during (60.seconds)))
  .protocols(httpProtocol)
  .assertions(
    global.responseTime.max.lt(2000),        // Max response time < 2s
    global.responseTime.mean.lt(500),        // Mean response time < 500ms
    global.successfulRequests.percent.gt(95) // Success rate > 95%
  )

πŸ“Š Running Tests​

Command Line​

# Run all simulations
./bin/gatling.sh

# Run specific simulation  
./bin/gatling.sh -s com.example.BasicSimulation

Maven​

mvn gatling:test

πŸ“ˆ Reports & Analysis​

Gatling generates detailed HTML reports including:

  • Timeline - Request distribution over time
  • Response Time Distribution - Percentiles and histogram
  • Active Users - Concurrent user count
  • Requests per Second - Throughput metrics

Reports location: results/simulation-timestamp/index.html

πŸ” Monitoring & Debugging​

Real-time Monitoring​

// Add during simulation
.exec { session =>
  println(s"User ${session.userId}: Current response time trend")
  session
}

Detailed Logging​

val httpProtocol = http
  .baseUrl("https://api.example.com")
  .header("Accept", "application/json")
  .logLevel(Trace)  // Enable detailed logging

πŸš€ Best Practices​

  1. Start Small - Begin with few users, gradually increase
  2. Monitor Resources - Watch CPU, memory on test machine
  3. Realistic Data - Use production-like test data
  4. Think Time - Add pauses to simulate real user behavior
  5. Environment Isolation - Use dedicated test environment

πŸŽ“ Next Steps​

  • Karate + Gatling Integration (coming soon)
  • API Security Testing (coming soon)
  • CI/CD Performance Testing (coming soon)

Load test like a pro! πŸš€