This project is a simple multithreaded TCP server implemented in Java using a thread pool. It accepts multiple client connections concurrently and responds with a greeting message. The server is tested with Apache JMeter for load and performance benchmarking.

• Uses ExecutorService with a fixed thread pool.
• Accepts multiple client connections concurrently.
• Responds with a custom message: Hello from server <client_ip>.
• Demonstrates basic multithreaded socket programming in Java.
• Load tested with Apache JMeter.

1. The server starts on port 8010.

2. A fixed-size thread pool (default 10 threads) is created using Executors.newFixedThreadPool().

3. For every incoming client connection:

   • A new task is submitted to the thread pool.
   • The server sends back a greeting message to the client.

ServerSocket serverSocket = new ServerSocket(port);
Socket clientSocket = serverSocket.accept();
threadPool.execute(() -> handleClient(clientSocket));

The handleClient() method sends:

toSocket.println("Hello from server " + clientSocket.getInetAddress());

• Java 8 or above
• Apache JMeter (for performance testing)

1. Clone the repository:

   git clone https://github.com/NirajSalunke/ThreadPool-Server.git
   cd ThreadPool-Server

2. Compile the server:

   javac Server.java

3. Run the server:

   java Server

You can test using Netcat or Telnet:

nc localhost 8010

Expected output:

Hello from server /127.0.0.1

The server was tested with Apache JMeter 5.6.3 using a TCP Sampler.

• Number of Samples: 13,829
• Average Response Time: ~1 ms
• Median Response Time: 1 ms
• Deviation: 2 ms
• Throughput: ~60,039 requests/minute
• Errors: 0

The graph shows stable throughput with very low latency, demonstrating the efficiency of the thread pool approach.

• Add configurable port and pool size via command-line arguments.
• Implement request/response processing instead of static messages.
• Add logging support.
• Provide a client implementation for easier testing.
• Containerize with Docker for deployment.

This project is licensed under the MIT License - see the LICENSE file for details.