Push Email - Comparison With Polling Email

Comparison With Polling Email

Regardless of whether the receiver uses polling email, outgoing mail is generally pushed from the sender to the final mail delivery agent (and possibly via intermediate mail servers) using Simple Mail Transfer Protocol. However, if the receiver uses a polling email delivery protocol, the final step from the last mail delivery agent to the client is done using a poll. Post Office Protocol (POP3) is an example of a polling email delivery protocol. At login and later at intervals, the mail user agent (client) polls the mail delivery agent (server) to see if there is new mail, and if so downloads it to a mailbox on the user's computer. Extending the "push" to the last delivery step is what distinguishes push email from polling email systems.

The reason that polling is often used for the last stage of mail delivery is that, although the server mail delivery agent would normally be permanently connected to the network, it does not necessarily know how to locate the client mail user agent, which may only be connected occasionally and also change network address quite often. For example, a user with a laptop on a WiFi connection may be assigned different addresses from the network DHCP server periodically and have no persistent network name. When new mail arrives to the mail server, it does not know what address the client is currently assigned.

The Internet Message Access Protocol (IMAP) provides support for polling and notifications. When a client receives a notification from a server, the client may choose to fetch the new data from the server. This makes retrieval of new messages more flexible than a purely push system, because the client can choose whether to download new message data.