Multithreading Technique

A Java-based demo project for handling high-concurrency external API requests using multithreading techniques such as ExecutorService, ForkJoinPool, and CountDownLatch.

This project simulates sending between 1 and 100,000 requests in parallel to an external API using a combination of fixed thread pools and the common fork-join pool, depending on the request load. Built using Spring Boot and RestTemplate, the logic is triggered periodically using @Scheduled.

Features

• Supports high-load simulations (1 to 100,000 API calls)
• Uses ExecutorService with dynamic thread pool sizing
• Automatically switches to ForkJoinPool.commonPool() for heavy loads (above 10,000 requests)
• Uses CountDownLatch to synchronize request completion
• Avoids overlapping scheduled executions using isRunning flag
• Uses Spring’s RestTemplate for external API invocation
• Logs all responses and errors per request

Technologies

• Java 17+
• Spring Boot
• Scheduled Tasks (@Scheduled)
• ExecutorService, ForkJoinPool
• RestTemplate
• CountDownLatch
• SLF4J Logger

How to Run

1. Clone the repository

   git clone https://github.com/sahuri/ThreadTest.git
   cd ThreadTest
   

2. Update the API endpoint

   Edit TestService.java and update this line to point to your desired API:

   private final String apiUrl = "http://localhost:8000/task/submit";

3. Adjust the number of requests

   Edit the determineRequestCount() method in TestService.java to control the request load:

   private int determineRequestCount() {
    return 1000; // Change this number as needed
   }