📊 Benchmarking & Performance Testing Guide

This project showcases an example Java web application (EAR-based) benchmarked using Apache HTTP Server’s ab, VisualVM, and Apache JMeter. Developers will learn how to set up the environment, execute load tests, and collect performance metrics.

---

📝 Overview

While the application is packaged as an EAR (Enterprise Archive) with standard modules (Core, EJB, Web), the primary focus here is benchmarking request performance under different tools. You'll learn:

1. How to download and install required software 🛠️
2. Where to unpack project files 📂
3. Load testing with Apache ab 🚀
4. Profiling with VisualVM 🔍
5. Advanced load testing with Apache JMeter 🎛️
6. Generating & analyzing JMeter results 📈

---

1. Download & Install Software

Tool	Download Link	Notes
Java JDK 11+	https://adoptium.net/	Install and set JAVA_HOME.
Apache HTTP Server (Apache24)	https://www.apachelounge.com/download/	Needed for ab (Apache Bench).
VisualVM	https://visualvm.github.io/download.html	For JVM profiling and monitoring.
Apache JMeter	https://jmeter.apache.org/download_jmeter.cgi	Extract and configure .jmx tests.

Steps:

1. Download the ZIP or installer for each.
2. Follow on-screen instructions or extract to a folder of your choice.
3. Ensure executables (java, ab, visualvm, jmeter) are on your PATH.

---

2. Project Setup & Extraction

1. Clone or download this repository:

   git clone https://github.com/Tharindu714/Perfomance-Testing-Guide-Project.git

2. Extract (if ZIP) to:

   C:\Projects\BenchmarkingExample
   

3. Deploy the EAR to your application server (e.g., GlassFish, Payara, WildFly):

   asadmin deploy C:/Projects/BenchmarkingExample/ear/BenchmarkApp.ear

4. Verify the web module is running (e.g., http://localhost:8080/app).

---

3. Load Testing with Apache Bench (ab)

Apache Bench is a simple CLI tool for HTTP request benchmarking.

# Single concurrency test:
ab -n 1000 -c 10 http://localhost:8080/app/api/hello

• -n 1000: Total requests.
• -c 10: Concurrent clients.

# High-load test with custom header:
ab -n 5000 -c 50 -H "Authorization: Bearer token123" http://localhost:8080/app/api/data

Typical Output Fields:

• Requests per second: Throughput.
• Time per request: Latency.
• Transfer rate: Bandwidth.

Save results:

ab -n 10000 -c 100 http://localhost:8080/app/api/hello > ab_results.txt

---

4. JVM Profiling with VisualVM

VisualVM helps visualize heap, threads, and CPU usage.

1. Launch VisualVM:

   visualvm

2. In Applications pane, locate your JVM (e.g., GlassFish Server).

3. Monitor:

   • CPU: Real-time CPU load.
   • Monitor: Heap usage and GC activity.
   • Threads: Thread counts and states.

4. Profile:

   • Switch to Sampler or Profiler.
   • Start a CPU or Memory profiling session.

5. Run load tests concurrently (e.g., using ab or JMeter) to see live metrics.

📋 Tip: Take snapshots before and after tests to compare.

---

5. Advanced Load Testing with Apache JMeter

1. Setup JMeter Test Plan

2. Analyze Text Log

3. Result Tree View

4. Summary Report View

5.1 Create or Load a Test Plan

• Open JMeter GUI:

  jmeter

• Design test plan:

  1. Thread Group: Set number of threads (users), ramp-up, and loop count.
  2. HTTP Request sampler: Configure server, port, and path.
  3. Listeners: Add View Results Tree, Summary Report, Aggregate Report.

• Save your test as benchmark_test.jmx:

  File ▶ Save Test Plan As ▶ benchmarking/benchmark_test.jmx
  

5.2 Run Tests in Non-GUI Mode

jmeter -n -t C:/Projects/BenchmarkingExample/jmeter/benchmark_test.jmx \
       -l C:/Projects/BenchmarkingExample/jmeter/results.jtl \
       -e -o C:/Projects/BenchmarkingExample/jmeter/reports

• -n: Non-GUI mode.
• -t: Test plan file.
• -l: Log file (.jtl).
• -e: Generate dashboard report.
• -o: Output folder for HTML report.

---

6. Generating & Viewing JMeter Results 📈

1. After test execution, open the HTML dashboard in your browser:

   C:/Projects/BenchmarkingExample/jmeter/reports/index.html
   

2. Review metrics:

   • Throughput: Requests per second.
   • Average & 90th percentile response times.
   • Error %: Failed requests.
   • Graphs: Response times over time.

3. Export specific tables as CSV for reports.

---

✅ Additional Tips

• Network Emulation: Use tools like tc (Linux) or Clumsy (Windows) to simulate latency/packet loss.
• Distributed Testing: Run JMeter in master-slave mode for large-scale tests.
• CI Integration: Automate benchmarks in Jenkins or GitHub Actions.
• Baseline Comparisons: Always compare against a known baseline to detect regressions.

---

Your benchmarking journey starts here! 🚀 Adapt these steps to your environment and share your findings. Good luck! Cheers! 🎉