Five principles for selecting a database server:

1) High-performance principle

To ensure that the server purchased, not only to meet the operation of the operating system and the need for business processing, but also to meet the growth in business volume over a certain period of time. Generally you can calculate the required server TpmC value based on empirical formula (Tpmc is a program that measures the transaction processing capacity of a computer system), and then compare the TpmC values published by various server vendors and TPC organizations to select the appropriate model. At the same time, the market price/offer price of the server is used to divide the calculated TpmC value to find out the price per unit of TpmC value, which in turn leads to the selection of servers with high performance-to-price ratio.

Conclusion: server processor performance is critical, the main frequency of the CPU should be high, there should be a large cache

2) Reliability principle

Reliability principle is the primary consideration in all the selection of equipment and systems, especially in the large-scale, with a large number of processing requirements, need to run for a long time on the system. Consider the reliability of the server system, not only to consider the reliability or stability of individual nodes of the server, but also to consider the overall reliability of the connection between the server and related auxiliary systems, such as: network systems, security systems, remote printing systems. Where necessary, consideration should also be given to the use of clustering technology for critical servers, such as: dual-machine hot backup or cluster parallel access technology, or even the use of possible fully fault-tolerant machines.

Conclusion: the server should have redundancy technology, at the same time, such as hard disk, network card, memory, power supply and such equipment should be stable and durable, the performance of the main second.

3) scalability principle

To ensure that the server purchased has excellent scalability principle. Because the server is the core of all system processing, requires a large data throughput rate, including: I / O rate and network communication rate, and the server needs to be able to deal with a certain period of business development brought about by the amount of data, the server needs to be able to corresponding time on its own according to the needs of the business development of the corresponding upgrades, such as: upgrading of the CPU model, expanding the memory, expanding the hard disk, replacing the network card, Increase the number of terminals, connecting disk arrays or cluster systems with other servers for concurrent access to centralized data. All of these require that the servers purchased have a good margin of expansion on the whole. Generally database and billing application servers in the design of large-scale billing system will be used to increase the reliability of the cluster approach, which is connected to the disk storage system, according to the amount of data and investment considerations, you can use DAS, NAS or SAN and other implementation techniques.

Conclusion: The IO of the server should be high, otherwise there will be bottlenecks when the CPU and memory are high performance. In addition, the server scalability should be good, in order to meet the needs of the enterprise in the future development.

4)Security Principle

Servers deal with mostly core data of the system in question, on which key transactions and important data are stored and run. These transactions and data for the owner is an important asset, their security is very sensitive. The security of the server is inextricably linked to the overall security of the system, e.g., the security of the network system, data encryption, cryptographic regimes, etc. The server needs to be in its own, including hardware and software, should be designed from the perspective of security considerations, with the help of external security facilities to protect, but also to ensure their own high security.

Conclusion: First of all, from the server's material to have high hardness and high protection and other conditions, and secondly, the server's cooling system and the ability to adapt to the environment should be strong, so as to be able to meet the requirements of the server security in hardware.

5) manageability principle

Server is both the core and the system as a whole of a node part, like the network system needs to be managed and maintained, but also need to carry out effective management of the server. This requires standard management system support from the server's hardware and software, especially the operating system on it, but also some important system components.

Conclusion: Try to choose a server that supports as many systems as possible, because the more systems the server is compatible with, the more choice you can have.