The Digital Research Initiative
The University of North Carolina at Chapel Hill

Part Four: How Do Internet Servers and Clients Work?

Internet servers work by responding to requests by clients. Generally, this response includes sending data back to the client, which then displays the data for the user.

The interaction between Internet clients and servers is the foundation for the usefulness of the Internet. It is through these exchanges that Electronic Mail (EMail) is sent from one user to the next, information and multimedia is retrieved from the World Wide Web, and files are shared via FTP. The different forms of exchanges are referred to as 'services.'

There are literally thousands of different services on the Internet, but there are only a few that are very popular. In addition to the ones mentioned above, popular services include:

  • Multimedia services, such as RealMedia, Windows Media, and QuickTime.
  • Peer-to-Peer file sharing, such as Napster, Kazaa, and Windows File Sharing
  • Real-time chat services, such as Microsoft Messager, AOL Instant Messaging, and SameTime Instant Messager
  • DNS resolution service
Each of these services has its own (sub-)protocol, which (as discussed in Part One) is a method for conducting the service. All of these protocols operate in a similar fashion, so we can use one analogy to summarize the activities for all. Let's compare the example of a consumer requesting and receiving a refund on a defective refrigerator with the example of a client computer requesting and receiving a Web page.

In order to request the refund on the defective refrigerator, the consumer would need to send a letter through the United States Postal Service. This letter will need to be specifically addressed to reach the Refunds Department of the manufacturer of the refrigerator, and the letter will need to clearly explain the needs of the consumer. This letter will be addressed as follows:

    Attn: Refund Department
    Refrigerator Maker
    813 Newton Street
    Pittsburgh, PA
In the letter, the consumer will explain that the refrigerator no long cools food, that the refrigerator is less than a year old, that the consumer is entitled to a refund according to the manufacturer's warranty, and that the consumer expects to receive this refund. Upon receiving the letter, the manufacturer will review the request, write a check for the refund amount and send the refund to the consumer.

Similarly, when a client computer (acting at a user's instruction) sends a request for a Web page to a Web Server computer, the computer has to send that request to the attention of the Web Server software that resides on the server computer. This is done by sending the request to a specific port (port 80). The client computer will need to specify which Web page it wishes to receive. This is represented in the "URL" (Uniform Resource Locator) of the Web page, which may look like this:

    http://www.unc.edu/depts/index.html
This address contains the protocol, the server computer's DNS Name, and the exact Web page the users wishes to view. These parts are:
  • http:// - This identifies the Web protocol, sent via port 80.
  • www.unc.edu - This identifies the Web Server computer by DNS Name.
  • /depts/index.html - This identifies the location of the Web page in the Web Server's long-term storage.
It is important to note that the Web Server's ability to answer this request is dependent upon it having a running software program that is designed to answer these requests. As mentioned in Part One, a computer only acts on instructions from a human being, and in this case, those instructions are in the request from the client, the client's Web browser software and the server's Web Server software.

Upon receiving the request for http://www.unc.edu/depts/index.html, the www.unc.edu Web Server will locate the file 'index.html' in the 'depts' directory, copy this file to its short-term storage, and then send this file back to the client computer. The client computer will receive the index.html file, and then display it on the client's monitor for the user to read and use the information in index.html.

The steps in these two examples relate to one another as follows:

Consumer/Refund Web Request
The consumer realizes the refrigerator is defective. The user decides to view the UNC Departments Web page.
The consumer sends a letter to the refrigerator maker requesting a refund. The user instructs the client computer to send a request to the www.unc.edu Web Server for the Departments Web Page.
The letter is delivered to the refrigerator maker according to the Postal Service procedure. The www.unc.edu DNS Name is resolved according to the procedure explained in Part Three, and the request is delivered to the www.unc.edu server according to the procedure explained in Part Two.
The refrigerator company reviews the request, writes the refund check, and sends the check back to the consumer. The Web Server processes the request, finds the requested file, and sends a copy of it to the client computer.
The consumer cashes the check at their bank and spends the money on a new refrigerator. The client computer receives the file, the client computer displays the file on its monitor, and the user reads the information.

Some important points to note about this procedure:

  • The request will only be answered correctly if it is prepared properly. If the consumer sends a letter filled with vulgarity, they will receive a dimissive letter from the refrigerator manufacturer instead of a refund. If the client computer does not use the appropriate Web protocol, the Web Server will respond with an error message instead of the desired file.
  • The request requires some activity from the server in order to be answered. The refrigerator manufacturer must review the request and write a check before it can answer the request. In this case, the Web Server must find and copy the appropriate file before sending the request back. In some cases, the Web Server may be required to complete a more complex procedure in order to send back an appropriate request, such as in the case of search engines.
  • By sending the request, the client is revealing a lot of information about itself that can be used in answering the request. The consumer reveals their street address, the kind of paper and color of ink they prefer, and their style of writing. The refrigerator maker can use this information to review/process the request for a refund or they could save this information for future studies on how consumers seek refunds. The client computer sends its address and its client software name/version. The Web Server may use this information to answer the requesting, particularly to create a version of the file that is best displayed by the client's software.
Of course, there are many unique qualities to the Web service, however, the procedure for requesting Web pages is very similar to the procedure for requesting any Internet service.

It should be noted that any computer on the Internet can be made into a server. All that is required is for the appropriate server software to be installed and configured onto the computer. Not all servers are equal, as there is a wide variance in the capabilities of computers on the Internet. Large Web sites, like www.unc.edu, are served by very powerful computers.

More information on Web Servers can be found at HowStuffWorks.Com.

Return to Part Three | Part Five: How Does Email Work?


This site made by
eric
This site was created especially for students of the UNC School of Journalism's
JOMC 050 Class, and anyone else who may be interested.
For more information, please contactdaikat@email.unc.edu