Application Layer of Computer Network

The application layer of Computer Network 1. Introduction The application layer serves as the highest level of interaction with users, and its tasks directly serve users' application processes. The specific content of the application layer is the protocol that the application process follows during communication. There are many application-layer protocols on the Internet, such as the most well-known HTTP (port 80) and FTP (port 21), SMTP (port 25) that supports email, DNS (port 53). Next we will briefly introduce Domain Name System DNS, World Wide Web (HTTP) and email (SMTP/POP3 ). 2. Domain Name System (DNS) is used to convert the names of machines that are easy to use to IP addresses. A domain name, separated by the decimal point, is a top-level domain name, second-level domain name, third-level domain name from right to left ...... Top-level Domain names (TLD) have 265 (2006.12 statistics ), such as cn/us/uk/jp, and com/net/org/edu/gov. Note that "abc" in www.abc.cn is higher than "abc" in www.abc.com.cn. The following describes the Domain name server ). There are four types of domain name servers: Root/top level/permission/Local Domain Name Server. The first three levels constitute the Domain Name Server: The Root Domain Name Server is the most important, at the top layer, stores the domain name and IP address of all top-level domain name servers; the top-level domain name server is responsible for managing all its second-level domain names. The next step is the permission Domain Name Server, which manages its own zone. Note that a domain can be divided into one or more zones, each of which is governed by a permission Domain Name Server. For example, a domain abc.com can be divided into a zone abc.com and a zone x.abc.com, this time domain abc.com is larger than the region abc.com, and the region abc.com management domain abc.com contains only x.abc.com. However, when a request is sent to x.abc.com, the domain name server with the permission to access zone abc.com is directed to the Domain Name Server with the permission of Zone x.abc.com, it is not a direct access to the x.abc.com permission Domain Name Server (because it will be resolved to the domain abc.com in order ). What does a Local Domain Name Server do? It is specialized for host services such as ISP/University/organization. The host's DNS query request will be sent to its local Domain Name Server for processing, and then the local domain name server will respond (recursive query-this server is solely delegated ). What is "recursion" reflected in? When the local domain name server does not know the answer, it will send a query request to the Root Domain Name Server (it must be the root). Everything is done by itself, and finally a host result is provided, "We are very conscientious and responsible ". This is different from the following. The Local Domain Name Server queries the Root Domain Name Server by iteration (iterativequery). To avoid excessive load, the Root Domain Name Server only tells the Local Domain Name Server which top-level domain name server to query, instead of taking care of the rest of the job. The Local Domain Name Server initiates the next query based on its "Directions. The rest are also iterative. In fact, the root domain name server does not only accept iterative queries-when the local Domain Name Server is "lazy", it will send recursive queries to the root domain name server, and the root server will still accept the full proxy. In general, iterative queries are used. Iii. World Wide Web (WWW) is a large-scale online information storage Institute (Web. Its soul is "Link", which connects resources on the Internet. The World Wide Web uses the Uniform Resource Locator URL (Uniform Resource Locator) to identify various documents on WWW, and uses the HyperText Transfer Protocol HTTP (HyperText Transfer Protocol) to implement links (using TCP Protocol ), use HyperText Markup Language (HTML) to display different documents in a unified manner. URL: <protocol >://< host >:< port>/<path> HTTP: transaction-oriented, stateless protocol, but reliable transmission is achieved using the TCP protocol. HTTP/1.0 is characterized by re-establishing TCP links and releasing TCP links for each document request, while HTTP/1.1 has improved, persistent connection used -- reserved for a TCP connection for a period of time. The HTML will not be mentioned. Generally, people who can see this article have at least a basic understanding, and the online tutorials are too rich, please use the World Wide Web Information Retrieval System (full-text retrieval and Classified Retrieval) on your own! 4. TCP connections are used for the Simple Mail Transfer Protocol and Post Office Protocol POP3. The former is used by the sender's User Agent to send mail to the sender's mail server and the sender's mail server to send mail to the recipient's mail server; the latter is used by the recipient's mail server to send mail to the recipient's UA. This is actually the figure below. It's over after understanding it. At last, I want to add more content. Because SMTP can only transmit printable 7-bit ASCII code, it is proposed that generic Internet Mail extended MIME (Multipurpose Internet MailExtensions) can transmit a variety of data, including multimedia. It does not replace SMTP, but sets a layer in the middle, such. POP3 also has a disadvantage, that is, it is set to delete the mail as long as the recipient UA reads the mail. Therefore, the Internet Message Access Protocol IMAP (Internet MessageAccess Protocol) is proposed, which not only does not actively delete, but also provides various management functions. It can be found that the current mailbox is IMAP.