Compare the advantages of blade server and rack server

&http://www.aliyun.com/zixun/aggregation/37954.html ">nbsp; The so-called Blade server (which is precisely called a blade server) means that multiple, modular server units can be plugged into a standard, rack-mounted chassis to achieve high availability and high-density. Each "blade" is actually a system motherboard. They can start their own operating systems via a "onboard" hard disk, such as Windows nt/2000, Linux, and so on, similar to a separate server in which each motherboard runs its own system, serves the different user groups specified, and is not associated with each other. However, administrators can use system software to assemble these boards into a single server cluster. In cluster mode, all of the motherboard can be connected to provide a high-speed network environment, and simultaneously share resources for the same user group services. By inserting a new "blade" into the cluster, you can improve overall performance. Because each "blade" is hot-swappable, the system can be easily replaced, and the maintenance time minimized.

These blade servers at the beginning of the design have low power consumption, small space, the price of a single machine, and it also inherits some of the traditional server's technical indicators, such as the use of hot Plug and redundancy in the blade server, these designs meet the dense computing environment for server performance requirements Some of them also improve the stability of the server and the core network performance through the built-in load balancing technology. From the appearance, compared with the traditional rack/tower server, Blade Server can maximize the use of the server space and cost, and provide users with flexible, convenient means of extending the upgrade.

The rack-type server does not look like a computer, and like a switch, it has 1U (1u=1.75 inches), 2U, 4U specifications. Rack-mounted servers are installed in a standard 19-inch cabinet. This structure is more of a functional server.

For information service enterprises (such as ISP/ICP/ISV/IDC), select the server first to consider the size of the server, power consumption, heat and other physical parameters, because information services enterprises usually use large private rooms to unify the deployment and management of a large number of server resources, the room is usually equipped with strict security measures, Good cooling system, multiple backup of the power system, the cost of its room is quite expensive. How to deploy more servers in a limited space is directly related to the cost of services to the enterprise, usually using a machine size to meet the 19-inch industrial standard rack-type server. Rack-type servers also have a variety of specifications, such as 1U (4.45cm High), 2U, 4U, 6U, 8U, and so on. Typically 1U rack-less servers are the most space-saving, but poor performance and scalability are appropriate for some of the business's relatively fixed areas of use. More than 4U of products with high performance, scalability, and generally support more than 4 high-performance processors and a large number of standard hot-swappable parts. Management is also very convenient, manufacturers usually provide people with appropriate management and monitoring tools, suitable for large access to key applications, but the volume is large, space utilization is not high.

We often see servers that can be divided into three types from the appearance type, namely the tower server, the rack server and the blade server. Because of the limited space of the enterprise room and other factors, blade server and rack servers are more and more popular with users, then what is the characteristics of the blade and rack server in the end which is better? This article will discuss this problem with you.

First, rack server and its characteristics

A rack server is a server that can be installed directly into a standard 19-inch cabinet. Typically such servers are somewhat similar to switches in size, so the rack server is actually a product of industry standardization, and its appearance is designed according to a uniform standard and is used uniformly with cabinets to meet the enterprise's server-intensive deployment requirements. The main role of the rack server is to save space, because the ability to put more than one server to a cabinet, not only can occupy a smaller space, but also easy to unified management. The height of a common cabinet is 42U (1u=1.75 inches or 4.4 centimeters), the rack server is 19 inches wide, and most rack servers are 1u-4u high.

The advantage of a rack server is that it takes up less space and facilitates unified management, but because of the internal space constraints, extensibility is limited, for example, 1U of servers are mostly only 1 to 2 PCI expansion slots. In addition, the thermal performance is also a need to pay attention to the problem, in addition to the need for organic cabinets and other equipment, so this server for more servers more large enterprises use, there are many enterprises using this type of server, but the server delivered to the dedicated server hosting agencies to host, In particular, many Web sites are currently using this approach to the server.

Second, blade server and its characteristics

Blade server is a highly available, high-density, Low-cost server platform is specially for the special application industry and high-density computer environment design, its main structure is a large body chassis, inside can insert many "blades", where each blade is actually a system motherboard, similar to a separate server, They can start their own operating system from the local hard disk. Each blade can run its own system, serving a specified set of different user groups, with no association to each other. Also, you can use system software to assemble these boards into a single server cluster. In cluster mode, all blades can be connected to provide a high-speed network environment, share resources, and serve the same user base. By inserting new blades into the cluster, you can improve overall performance. Because each blade is hot-swappable, the system can easily be replaced and the maintenance time minimized.

Depending on the server capabilities required, the blades are divided into server blades, network blades, storage blades, management blades, Fibre Channel SAN blades, extended I/O blades, and the like blades for different functions. Blade server has recognized two characteristics, one is to overcome the shortcomings of the chip server cluster, has become the Terminator of the cluster, the other is the implementation of the Cabinet optimization.