The UDP and TCP Headers Compared

Consider the UDP and TCP headers. First, UDP is a much simpler protocol than TCP, and so the fact that UDP's header is simpler than TCP's should be no surprise. The material below compares the two headers, points out similarities, and explains the reasons behind the differences. To the right are images of the two headers as found at http://cs.uccs.edu/~cs522/msgformat/format.htm.

• UDP is connectionless, but TCP is connection-oriented. Thus, TCP has fields used for connection maintenance, specifically the flags. UDP does not.
• Both protocols have the same port number fields, and they are used the same way in each protocol.
• TCP uses acknowledgements to provide reliable transport, whereas UDP does not provide reliable transport. So TCP needs a sequence number and acknowledgement number, but UDP does not.
• Because of the options, TCP header lengths vary. Thus, TCP needs the header length field to allow the receiver to separate the end of the header from the data. UDP has no options, no need for a length field, and no need to pad the options field out to a multiple of 32 bits in size.
• TCP supports flow control, but UDP does not. TCP's window size field (rwnd in the text, or Window in the original specification) is used for this. UDP has no analog.
• Both protocols use checksum in like manners, and so both have identical checksum fields.
• UDP's length field is necessary in order for the receiving process to have access to the size of the incoming datagram. TCP is a byte-stream protocol, and all bytes are the same length, so a length field is unnecessary.

Back to the IS 450/650 home page.