JMeter Load Testing Tutorial

Configure JMeter for 500 RPS Load Testing

🧮 RPS Calculator

Calculate the optimal JMeter settings for your target RPS:

Understanding Load Testing Fundamentals

Key Formula:
RPS = (Number of Threads × Loop Count) ÷ (Test Duration + Think Time)
Important: Achieving exactly 500 RPS requires careful balance of all parameters. The actual RPS depends on server response time, network latency, and JMeter's ability to process responses.

🎯 What is RPS?

Requests Per Second (RPS) measures how many HTTP requests your system can handle in one second.

500 RPS means your system should process 500 requests every second during the test.

⚡ Why 500 RPS?

This is a significant load that can reveal:

  • Database connection limits
  • Memory usage patterns
  • CPU bottlenecks
  • Network saturation points

JMeter Parameters Deep Dive

👥 Number of Threads (Users)

What it is: The number of virtual users JMeter will simulate concurrently.

How it affects RPS: More threads = potentially higher RPS, but diminishing returns apply.

For 500 RPS: Start with 50-100 threads. If response time is 200ms, you need fewer threads than if response time is 2 seconds.
Optimal Threads ≈ Target RPS × Average Response Time (in seconds)

📈 Ramp-up Period

What it is: Time taken to start all threads gradually.

Why it matters: Prevents overwhelming the server with sudden load spikes.

Recommended: 1-2 minutes for 500 RPS tests. This gives time to identify bottlenecks as load increases.
Thread Start Rate = Number of Threads ÷ Ramp-up Period

🔄 Loop Count

What it is: How many times each thread executes the test plan.

Options:

  • Infinite: Threads run until test duration ends
  • Fixed: Each thread runs exactly N times
For sustained 500 RPS: Use "Infinite" with a duration-based test controller.

⏱️ Test Duration

What it is: Total time the test runs.

Recommendations:

  • 5-10 minutes: Quick validation tests
  • 30-60 minutes: Stress testing and monitoring
  • 2+ hours: Endurance testing

⏳ Think Time (Timer)

What it is: Pause between requests to simulate real user behavior.

Impact on RPS: Longer think time = lower RPS from same number of threads.

Effective RPS = Threads ÷ (Response Time + Think Time)
For pure load testing: Use 0ms think time
For realistic simulation: Use 1-5 seconds think time

Step-by-Step Configuration

🛠️ Basic Test Plan Setup

  1. Create Test Plan
    Right-click Test Plan → Add → Threads → Thread Group
  2. Configure Thread Group
    • Number of Threads: 100
    • Ramp-up Period: 60 seconds
    • Loop Count: Infinite ✓
    • Duration: 300 seconds
  3. Add HTTP Request
    Right-click Thread Group → Add → Sampler → HTTP Request

🎛️ Advanced Settings

Constant Throughput Timer

Location: Thread Group → Add → Timer → Constant Throughput Timer
Target Throughput: 30000 (samples per minute for 500 RPS)
Calculate based on: All active threads

HTTP Request Defaults

Location: Thread Group → Add → Config Element → HTTP Request Defaults
Settings:
  • Server Name: your-api-server.com
  • Port: 443 (HTTPS) or 80 (HTTP)
  • Protocol: https or http

📊 Essential Listeners

  • Summary Report: Overall test statistics
  • Aggregate Report: Detailed response time metrics
  • View Results Tree: Individual request/response details (disable during load test)
  • Simple Data Writer: Export results to CSV
Performance Tip: Disable View Results Tree and other GUI listeners during actual load testing. They consume significant resources and can affect test accuracy.

Monitoring and Optimization

📈 Key Metrics to Watch

  • Throughput: Actual RPS achieved
  • Response Time: Average, 90th, 95th percentiles
  • Error Rate: Should be < 1% for valid tests
  • Active Threads: Number of concurrent users

🔧 Troubleshooting Common Issues

Not Reaching 500 RPS?

  • Increase thread count
  • Reduce think time
  • Check JMeter heap size (-Xmx4g)
  • Verify server capacity

High Error Rates?

  • Reduce ramp-up speed
  • Check connection timeouts
  • Verify server resources
  • Review application logs

⚙️ JMeter Performance Tuning

JVM Arguments:
-Xms1g -Xmx4g -XX:+UseG1GC
-XX:MaxGCPauseMillis=100
-Djava.net.preferIPv4Stack=true
JMeter Properties:
jmeter.save.saveservice.output_format=csv
jmeter.save.saveservice.response_data=false
jmeter.save.saveservice.samplerData=false
Hardware Considerations: For 500 RPS testing, ensure your JMeter machine has adequate CPU (4+ cores) and RAM (8+ GB). Consider distributed testing for higher loads.