Internet host using sockets and ports.

Sockets and Ports

TCP multiplexes multiple connections to a single Internet host using sockets and ports.

A socket is a network communications endpoint. The analogy is to a wire (the network data connection) being plugged into a socket.

Sockets come in two primary flavors. An active socket is connected to a remote active socket via an open data connection. Closing the connection destroys the active sockets at each endpoint. A passive socket is not connected, but rather awaits an incoming connection, which will spawn a new active socket.

A socket is not a port, though there is a close relationship between them. A socket is associated with a port, though this is a many-to-one relationship. Each port can have a single passive socket, awaiting incoming connections, and multiple active sockets, each corresponding to an open connection on the port.

bandwidth by performing flow control

TCP manages limited network bandwidth by performing flow control


Modern data networks are designed to support a diverse range of hosts and communication mediums. Consider a 200MHz Pentium-based host transmitting data to a 25MHz 80386/SX. Obviously, the Pentium will be able to drown the slower processor with data. Likewise, consider two hosts, each using an Ethernet LAN, but with the two Ethernets connected by a 28.8 Kbps modem link. If one host begins transmitting to the other at Ethernet speeds, the modem link will quickly become overwhelmed. In both cases, flow control is needed to pace the data transfer at an acceptable speed.

Request/reply flow control requires each data packet to be acknowledge by the remote host before the next packet is sent. Sliding window algorithms, used by TCP, permit multiple data packets to be in simultaneous transit, making more efficient use of network bandwidth. Finally, Internet's Unreliable Delivery Model allows packets to be discarded if network resources are not available, and demands that protocols make provisions for retransmission.

Sliding Window

TCP's method of flow control is called sliding window.

Sliding window algorithms are a method of flow control for network data transfers. TCP, the Internet's stream transfer protocol, uses a sliding window algorithm.

A sliding window algorithm places a buffer between the application program and the network data flow. For TCP, the buffer is typically in the operating system kernel, but this is more of an implementation detail than a hard-and-fast requirement.

Data received from the network is stored in the buffer, from whence the application can read at its own pace. As the application reads data, buffer space is freed up to accept more input from the network. The window is the amount of data that can be "read ahead" - the size of the buffer, less the amount of valid data stored in it. Window announcements are used to inform the remote host of the current window size.

If the local application can't process data fast enough, the window size will drop to zero and the remote host will stop sending data. After the local application has processed some of the queued data, the window size rises, and the remote host starts transmitting again.

On the other hand, if the local application can process data at the rate it's being transferred, sliding window still gives us an advantage. If the window size is larger than the packet size, then multiple packets can be outstanding in the network, since the sender knows that buffer space is available on the receiver to hold all of them. Ideally, a steady-state condition can be reached where a series of packets (in the forward direction) and window announcements (in the reverse direction) are constantly in transit. As each new window announcement is received by the sender, more data packets are transmitted. As the application reads data from the buffer (remember, we're assuming the application can keep up with the network), more window announcements are generated. Keeping a series of data packets in transit ensures the efficient use of network resources.