Network diagram for UDP

About this network diagram for UDP

In a UDP network, the sender and receiver communicate without a set connection. When the sender has data, it packages it into a datagram. This datagram has a simple header and a payload. The header includes the source port, destination port, length, and checksum fields.

Once ready, the sender quickly sends the datagram onto the network. They don’t wait for permission. The network forwards the datagram to the receiver using routing rules. Since UDP doesn't keep connection states, each datagram is independent. It may arrive out of order or not at all.

The receiver watches the specified port for incoming datagrams. When a UDP datagram arrives, it checks the length and the optional checksum if present. If the datagram passes these checks, the receiver accepts the payload. Then, it sends it to the correct application. The sender doesn’t receive a confirmation of delivery. There is no automatic acknowledgment. This method reduces delay and avoids the overhead of connection management.

UDP does not create or track session states. It also ignores sequence numbers, retransmission timers, and flow-control windows. Each datagram is generated, processed, and delivered without reference to previously sent packets. If the network becomes busy or datagrams arrive too fast, newer packets can overwrite older ones or may be discarded. The sender doesn’t know about these events. This is because UDP is designed for speed and simplicity, not reliability.

Applications using UDP usually manage error recovery and ordering themselves. UDP is perfect for real-time data, like streaming or voice. Here, timely delivery is more important than perfect accuracy. UDP provides fast and efficient data transfer. It does this with little internal processing.