In theory, a single communication, such as a music video or an email message, could be sent across a network from a source to a destination as one massive, uninterrupted stream of bits. If messages were actually transmitted in this manner, it would mean that no other device would be able to send or receive messages on the same network while this data transfer was in progress. These large streams of data would result in significant delays. Further, if a link in the interconnected network infrastructure failed during the transmission, the complete message would be lost and have to be retransmitted in full.

A better approach is to divide the data into smaller, more manageable pieces to send over the network. This division of the data stream into smaller pieces is called segmentation. Segmenting messages has two primary benefits:

The downside to using segmentation and multiplexing to transmit messages across a network is the level of complexity that is added to the process. Imagine if you had to send a 100-page letter, but each envelope would only hold one page. The process of addressing, labeling, sending, receiving, and opening the entire 100 envelopes would be time-consuming for both the sender and the recipient.

In network communications, each segment of the message must go through a similar process to ensure that it gets to the correct destination and can be reassembled into the content of the original message, as shown in Figure 2.

Various types of devices throughout the network participate in ensuring that the pieces of the message arrive reliably at their destination.