UDP stands for User Data Protocol, one of the key protocols of the Internet Protocol Suite. The other common protocol is the TCP. Programs on networked computers use UDP to send short messages known as datagrams to one another. The service provided by this protocol is not regarded as sufficiently reliable as it provides no guarantee for delivery or safety from duplication. Still, the datagrams may arrive out of order or go missing without a notice. Typically, the UDP takes messages from the application process, attaches source and destination port number fields for the multiplexing/demultiplexing service. It may then add two or more elements of marginal importance and passes the segment onto the. The layer encapsulates the segment into an IP datagram and then makes the “best effort” to try to deliver that segment to the intended recipient. If the segment arrives at the recipient, the UDP then utilizes the port numbers and the IP source and the destination address to deliver the data in the segment to the appropriate application process. It is to be noted that here, there is no “handshaking” between sending and the receiving transport-layer entities before sending a segment. One of the application layer protocols that use UDP is the Domain Name Server (DNS). When a DNS application in a host wants to make a query, it constructs a query; it makes the query message which it passes to the UDP socket. Without performing any handshaking, the UDP adds a header field fields to the data and passes on the segment to the network layer. The segment is then encapsulated into a datagram and the datagram is then sent to the name server. He DNS application then waits for the reply. If no reply is forthcoming (especially if the UDP lost the query on the way) , the DNS could then try to send the same query to another nameserver, or it informs the invoking application that it cannot ge a reply. it may then be difficult to justify a developer building an application over UDP rather than TCP, since the UTP is apparently more reliable. But it turns out that UDP in fact is more suited for some application for the following reasons. • There is no connection establishment. TCP on its part uses a three-way handshake before it starts to transfer data. But UDP proceeds to transfer data without any formalities and so there is minimal delay. Thus if DNS were to run over TCP, it would be much slower. • No connection state. TCP maintains a connection state in the end systems. The connection state includes receive and send buffers, congestion control parameters, and sequence and acknowledgement number parameters. UDP does not nee to maintain connection state and a server devoted to a particular application can support many more clients when it is running over UDP. • Small segment header overhead. With only about 8 bytes of overhead, UDP is lighter to transport more than the 20 bytes of overload for the TCP. • Unregulated send rate. TCP has a congestion control mechanism that restricts the sender when one or more links between sender and receiver becomes overwhelmingly congested. In the case of the UDP, the data is only constrained by the rate at which the application generates data, the capabilities of the source (CPU, clock rate, etc.) and the access bandwidth to the Internet.
Elizabeth Brown writes on topics such as UDP , OSI Model and Physical Layer for The Tech FAQ.
Article Directory: http://www.articlecrash.com/
Please Rate The Above Article From The Internet Category Article Title: What Is Udp?
5 out of 54 out of 53 out of 52 out of 51 out of 5
Not yet Rated
Syndicate Internet Related Articles Via RSS!
Copyright © Article Crash All rights protected. Use of our free service is protected by our Privacy Policy and Terms of Service
Powered by Article Dashboard