Skip to content

Latest commit

 

History

History
107 lines (77 loc) · 5.25 KB

README.md

File metadata and controls

107 lines (77 loc) · 5.25 KB

PetClinic JMeter Performance Test

Overview

This JMeter test plan is designed to benchmark and measure the scalability of the Spring PetClinic REST API. The test performs CRUD operations on the API endpoints, simulating real-world usage patterns.

What This Test Covers

  • Creating, updating, retrieving, and deleting owners and pets.
  • Scheduling and managing vet visits.
  • Configurable workload parameters for scalability testing.

Running the Test

Prerequisites

Ensure you have the following installed:

  • Apache JMeter 5.6.3+ (Download here)
  • Spring PetClinic REST API running locally 
    # Navigate to the Spring PetClinic REST project directory and start the application
cd /path/to/spring-petclinic-rest
mvn spring-boot:run

(Runs on http://localhost:9966 by default)

Running the Test from CLI

Run the JMeter test in non-GUI mode:

jmeter -n -t src/test/jmeter/petclinic-jmeter-crud-benchmark.jmx \
 -Jthreads=100 -Jduration=600 -Jops=2000 -Jramp_time=120 \
 -l results/petclinic-test-results.jtl

CLI Parameters

Parameter Description Default Value
threads Number of concurrent users 50
duration Duration of the test (seconds) 300
ops Target throughput (operations/sec) 1000
ramp_time Time to ramp up threads (seconds) 60

Analyzing Test Results

  1. Generate an HTML Report

    To generate an interactive HTML performance report:

    jmeter -g results/petclinic-test-results.jtl -o results/html-report

    Then open results/html-report/index.html in your browser.

  2. View Raw Results in CLI

    To quickly inspect the last 20 results:

    tail -n 20 results/petclinic-test-results.jtl

  3. Process CSV Results for Custom Analysis

    JMeter results are stored as .jtl files, which use CSV format by default unless configured otherwise. You can analyze them using:

    • Python Pandas: 
      import pandas as pd
df = pd.read_csv('results/petclinic-test-results.jtl')
print(df.describe())
    • Command-line tools (awk): 
      awk -F',' '{print $1, $2, $3}' results/petclinic-test-results.jtl | head -20

Performance Summary Table

Transaction Total Requests Avg Response Time (ms) Min Response Time (ms) Max Response Time (ms) 90th Percentile (ms) 95th Percentile (ms) 99th Percentile (ms)
Add Pet to Owner 60,284 60.17 1 534 151.00 190.00 255.15
Create Owner 60,284 145.94 2 696 315.00 368.00 476.15
Delete Owner 60,224 158.76 2 825 322.00 375.00 490.00
Delete Pet 60,239 179.94 2 829 358.00 411.00 550.00
Get Pet Belonging to Owner 60,249 70.37 1 479 171.00 208.00 260.50
Schedule Visit 60,279 166.48 2 880 329.00 381.00 507.20
Update Owner 60,264 163.76 2 681 345.00 396.00 486.35
Update Pet 60,249 33.53 2 402 84.00 108.50 161.50

Explanation:

  • Total Requests: Number of executions of the request.
  • Avg Response Time (ms): Average time taken for the request.
  • Min/Max Response Time (ms): The shortest and longest recorded response times.
  • Percentile Metrics: The 90th, 95th, and 99th percentile response times show performance under load.

Next Steps

  • Run with different configurations to simulate varied workloads.
  • Integrate into CI/CD pipelines for automated performance testing.

Contribute: Feel free to submit PRs to improve the test coverage or parameters!