Select the appropriate server for your own

Choosing a server is a major decision, usually time-consuming and expensive. Wrong decisions can lead to undesirable consequences. In this article, we'll start with some of the issues that need to be considered in planning, and then further discuss how to make a choice between Windows and UNIX.

Planning Considerations

A key consideration when selecting a server is determining what operating system to use. Here are a few things to consider:

Does the server want to increase to an existing server/computer environment?

What are the limitations of OS selection for existing or selected applications?

Adding a server to an existing environment usually means that the new server must use the same operating system as the existing server because of the limitations of the application and the limited skills of the existing operators.

If new applications are required, software vendors may require or strongly recommend the use of specific OS.

Windows vs. Unix and Linux

Unless you have to use some proprietary OS (such as some IBM system OS/40 or z/OS, or even Novell's NetWare), you typically have three options: Windows, Linux, or UNIX. UNIX includes all UNIX systems offered by system vendors, such as IBM's AIX, HP's Hp/ux, or Sun's Solaris. Similarly, Linux systems have some options.

The first option is to use Windows, or the extended family of Unix or Linux. As mentioned above, sometimes you may be forced to use only some kind of operating system, but if you are free to choose, it is best to compare products based on standards of scalability, robustness, and cost of use.

How many client computers must the server support?

The first basic step in server selection is to determine how many clients the server must support at the same time. Different applications use different server resources, and this information is needed when configuring the server.

Regardless of the number of these clients and applications, it is only the initial number. Over time, the number of business activities has increased, or computer-based activities have increased. This natural growth makes the system more important for extensibility.

Scalability is a measure of the system's ability to upgrade in different ways, including processing power, storage capacity, main memory size and network connectivity and bandwidth. The system that can be extended is a much cheaper and less intermittent option than replacing a larger system.

Scalability is critical, but how to reasonably estimate the extensibility of the server system, there is not much information, there is no practical experience. Moreover, there is no established test standard to measure extensibility.

Note that it is important that the OS that has been tested, tested, and tested for a long time has better functionality. In particular, multiprocessor operating systems, the longer the history of the operating system, often than the new system to provide better features.

What applications must the

support?

When considering which OS is best for your application, it's a good idea to break these into four categories:

file, print, or communication Server

Database server

Application Server

Compute-intensive services Device

This classification does not mean that a separate application cannot exist on the same server at the same time. However, systems of different uses can easily reach their limits because it is difficult to prioritize resource allocations among different categories of applications, unless the resource management program is used.

To evaluate the client's resource usage, first figure out which clients will use each application. Each application may require a different combination of resources for the server.

Work Properties

The next issue to be explained is the nature of the server support. Will it be used for business-critical applications, workgroup, departmental or enterprise-wide applications, or
other applications?

A server can be dedicated to a certain number of applications within a workgroup, a department, or the entire company. Obviously, deploying all the required applications to the same server seems to save costs, but this increases the vulnerability of the system because once an application crashes the system, all applications cease to work.

As we mentioned before, it is difficult to balance the resource needs of each application when a server is running uneven workloads. Although there are applications that can solve this resource allocation challenge, the problem is difficult to solve.

Generally, it is best to deploy multiple servers, each dedicated to one or a small number of activities. This reduces the vulnerability of the system, because if you have more than one server, you can move your work to another machine if a single server fails. The use of server redundancy is the foundation of a high-availability system.

Financial Issues

While the cost of purchasing servers and necessary software is clearly a big part of total costs, we also strongly recommend that total cost of ownership (TCO) be evaluated when making a purchase decision.

TCO includes both direct and indirect costs.

The main direct costs include:

The cost of purchasing software and hardware

Operation and Management system

Application development, support, and communications

Power costs for operating and cooling systems

System occupancy Space Cost

Indirect costs are more difficult to quantify, including the cost of system downtime. Downtime costs include not only the cost of handling system availability (for example, loss of productivity), but also the soft cost of being unavailable. As a result of system downtime, customers or suppliers suddenly unable to do business with you may turn to your competitors.

When the system is down or the reaction is slow, the user may not be able to do anything but wait for the system to recover. More often, they may be able to solve the problem by talking to each other and asking for help. In either case, they spend time and resources trying to understand what to do, which increases the company's costs.

Comparison of system price and performance is a benchmark (benchmark) to follow. But since these benchmarks are used to measure the server in a specific way, your usage is likely to be different. Although the benchmark number of results is true, it must be noted when used. It is recommended that these benchmarks be used only as an indication, or for a wide variety of possible screening. Do not rely on these benchmarks alone to make a purchase decision.

A system with better scalability is much more expensive than a system that does not have extensibility. However, when your server is unable to run the necessary workloads, it is often cheaper to purchase more computing power than you can upgrade the replacement system.