Virtual Hosting Basics

1. What is virtual hosting, and why use virtual hosting

Using virtual hosting technology, a real host can be divided into many "virtual" hosts. Each virtual host has an independent domain name and IP address, and has complete Internet server (www, FTP, email) functions. Virtual hosts are completely independent from each other. From the outside, each virtual host looks exactly the same as an independent host. The effect is the same, but the cost is very different. Because multiple virtual hosts share the resources of one real host, the hardware costs, network maintenance costs, and communication line costs borne by each virtual host user are greatly reduced, making the Internet truly a network everyone can afford! 

At present, many enterprises use this method to build websites. This not only greatly saves the cost of purchasing machines and leasing dedicated lines, but also eliminates the need to worry about the technical problems of using and maintaining servers, and even more, there is no need to hire dedicated administrators, because we take care of all these things.

If the resources of one machine are allocated to multiple virtual hosts, will the performance of each virtual host decline?

This problem depends on the performance of the host and the number of virtual hosts set up. If configured properly and high-bandwidth lines are used, virtual hosts will be faster than independent hosts connected via lower-bandwidth lines. We will control the number of users per host within a certain range, so as to make full use of host performance while safeguarding user interests.

2. What is server hosting

When a user intends to have his own Web, Email, FTP and other servers, he can choose two methods: one is self-building, and the other is hosting.
Self-building requires relatively high-level professional and technical personnel, a large investment in purchasing software and hardware, and also payment of daily maintenance and line communication fees with no upper limit, and the construction period is also relatively long. In comparison, hosting is economical, fast, and practical.
The basic operating mechanism of the Internet is client/server (Client/Server), and its greatest feature is crossing time, space, and regions. That is, even if people are far apart, as long as everyone is on the network, both parties can communicate directly. This provides us with an important technical means: remote control (RemoteControl). The technical basis and means on which server hosting depends are mainly mechanisms such as "client/server" and "remote control". That is, no matter where our users are, as long as they can access the Internet, they can control a server far away, thereby realizing ownership and maintenance of the server.

3. What is the difference between server hosting and virtual hosting

Server hosting services can be subdivided into server hosting and virtual hosting (Virtual Hosting). "Server hosting" means placing a server at a company with a network environment connected to the Internet in real time, or renting a server from it. Customers can configure the server into a WWW, E-mail, Ftp server through remote control. "Virtual hosting" means: dividing the hard disk of an entire UNIX or NT system machine into smaller parts. Each divided hard disk space can be configured as a www, email, FTP server with an independent domain name and IP address. When people browse such a server, they cannot tell that it shares the system resources of one host with others. Users who rent space on this machine can, through remote control technology such as file transfer (FTP), fully control the space belonging to them, such as uploading and downloading information, configuring application functions, etc. It should be said that owning an independent site through "virtual hosting" has a performance-price ratio far higher than building and maintaining a server by yourself. At present, this method of building a site is being adopted by more and more enterprises and institutions.

4. From which angles should server performance be evaluated

A. Server hardware configuration
B. Server software configuration
C. Operating system and development environment installed on the server
D. Functions supported by the server
E. Network environment where the server is located

5. What factors determine the speed at which a server is accessed

A. Server hardware configuration (including server type, CPU, hard disk speed, memory size, network card speed, etc.)
B. Network environment and speed where the server is located
C. The speed of the connection between the network environment where the server is located and the Internet backbone
D. The international gateway speed of ChinaNet
E. The leased line speed between the visitor's ISP (Internet access service provider) and ChinaNet
F. The port access speed opened by the visitor's ISP (Internet access service provider) to the client
G. The configuration of the visitor's computer, Modem speed, telephone line quality, etc.

6. Why provide Web+EMail+FTP server hosting at the same time

"Web, E-mail, FTP" functions are three relatively basic and commonly used functions of the Internet. The Web function enables visitors to enter a domain name through a browser and browse the homepage of the site owner, achieving the purpose of publicizing the company's image and products. The Email function, by setting up Email mailboxes with the company's own domain name, can both establish an image and create convenient, fast, and low-cost modern communication links. The FTP function can allow visitors to freely download files they are interested in, increasing the site's visit rate. These three functions are indispensable whether for users to publicize the company's image and products, or for after-sales service and online technical support.

7. Can multiple domain names point to the same website

By configuring the server, we can make multiple domain names point to the same website. Users can use multiple domain names to access the same homepage content on the same site.

8. How to test the speed of a virtual host

If you are directly connected to the network through a Modem or leased line, or if you need to use the Telnet function of a gateway, you can directly "ping" the network you want to know about. The PING command makes your current host send a "data packet" to the target host. When the target host receives this "data packet", it will immediately and automatically return a data packet. At this time, what you get is the time for one round trip of a data packet.
For example, "pingwww.microsoft.com", then press Enter, and you will get the response time for you to access Microsoft's server from China. Generally, the "PING" command will continue automatically until you terminate it, such as by pressing "ctrl + C".
If you are familiar with using the PING function in Telnet, and you have Telnet accounts on different networks, you can use different networks as a basis to perform PING tests on other hosts. You can compare the response times of these accessed hosts at different access points, and obtain test results such as "packet loss rate" and average response time.
However, the PING command can only test the relatively low-level performance of the network and server. It can reflect the bandwidth of network connectivity and the degree of data congestion, but it cannot well reflect higher-level server performance—among which the most important, such as the efficiency of the Internet Web server software and hardware running on the host (this efficiency directly determines the speed of file download through the Web server—that is, the speed at which users access the server), cannot be given. In fact, there are some well-known neutral organizations internationally that, in order to help users understand server performance and other server information, have opened some professional testing sites for users to perform these tests. Among them, the NetMechanic website run by Monte Sano Software provides users with free testing services for the main performance of servers.
The URL of this service website is: 　　　
　　 http://www.netmechanic.com/server_check.htm
Visit http://www.netmechanic.com/server_check.htm, and you can directly enter the domain name address of the server to be tested in the interface, and enter your own EMAIL address. In this way, the testing service will send a test to the target website every 15 minutes, and after 8 hours it will deliver the test results to your E-mail mailbox. This service can provide data and average values for each test point for the server's main operating performance, such as PING response time, DNS resolution time, CONNECT time, and Web server file download speed.
Now, through these methods, you can test your own website, etc., or other websites you are interested in. At the same time, you can test and compare Microsoft microsoft.com, Netscape netscape.com, yahoo.com, etc.
Because this service website is located in the United States (the ServerPro service also provides access tests originating from Europe), the test results use the United States as the access source. Therefore, we will find that this website shows very high network speed in test results for hosts located in the United States. This shows the access effect when a user located in the United States (or Europe) accesses our host. For test results of websites in other countries, the speed is relatively much lower—this is the network bottleneck problem of China's international gateway of data lines that we often hear about.
P.S. The Netmechanic free server performance testing website mentioned in the article may recently, due to excessive testing demand, encounter the situation of "Sorry, the testing quota is full. Please submit your application again later." As long as you try again after a while, you will definitely be able to join the free testing ranks, and you can enter multiple tests at the same time.

9. What is NT load balancing service

WLBS (Windows NT Load Balancing Service) is a network service used to manage TCP/IP communication between different servers. WLBS is not a fancy pseudo-application distributor used to share the load of server CPUs. It is more appropriately described as a more flexible dynamic load balancing service based on network traffic. The most commonly used load-sharing technology is Round Robin DNS (RRDNS). In RRDNS technology, requests are simply sent cyclically to different servers. Therefore, when one of these servers crashes, requests are still sent to this machine. WLBS determines whether the host server is working and whether it can accept requests, thereby overcoming this weakness. If the above situation occurs, WLBS transfers network communication to another server, thereby avoiding failures. 　 　　
When applications and Web sites need to meet the demands of growing traffic, you may even want to offload CPU from one server to another. Not long ago, you could only share the system load by choosing Round Robin DNS and expensive hardware. But now, WLBS for Microsoft Enterprise Edition Server 4.0 can help you distribute TCP/IP traffic on servers. This distribution service allows you to handle the growing load of a site by adding Web servers. 　 　　

To improve performance, the WLBS service provides real-time fault tolerance so that servers can be added and removed during maintenance. If you plan to add stronger functionality and better reliability to mission-critical Web applications, then the combination of the features described above constitutes an indispensable service.