Blade Server and disk array card (RAID) Technology

Recently, the customer needs to change the server. The customer sent the purchased server to a very small server. But when I saw it, it was a big, long, wide server, similar to the server in the data center, later, Mr. Mi gave a general explanation: This is a blade server. A blade server is a low-cost server platform that supports HAHD (HighAvailabilityHighDensity, high availability and high density ).

Recently, the customer needs to change the server. The customer sent the purchased server to a very small server. But when I saw it, it was a big, long, wide server, similar to the server in the data center, later, Mr. Mi gave a general explanation: This is a blade server.

A blade server is a low-cost server platform that can be inserted with multiple card-type server units in a standard-height rack-mounted chassis to achieve HAHD (High Availability and High Density, designed for special application industries and high-density computing environments. Like a blade, each blade is actually a system motherboard.

This is a simple understanding, cherish the opportunity, and learn from this knowledge. I am afraid of knowing the problem when I start to learn materials and solve it. what impressed me most is: when the customer came to me, he told me that the server had already been RAID, but I was a bit blind. If I didn't know what this was, I simply understood it as READ, only by looking for such information can we know that RAID is the real thing.

RAID is the abbreviation of Redundant Array of Independent Disks ).

To put it simply, RAID is an independent hard disk (physical hard disk: After observation, the blade server provided by the customer has four hot-swappable hard disks, each of which is 500 GB) A hard disk group (logical hard disk) is formed in different ways to provide higher storage performance than a single hard disk and Data Backup technology. RAID Levels is used to form disk arrays ). The data backup function is used to restore damaged data once the user data is damaged, thus ensuring the security of user data. In the user's view, a disk group is like a hard disk, which can be partitioned, formatted, and so on. In short, the operations on the disk array are the same as those on a single hard disk. The difference is that the storage speed of the disk array is much higher than that of a single hard disk, and automatic data backup can be provided.

Two features of RAID technology:

First, speed

Second, security

Thanks to these two advantages, RAID technology was applied to hard drive systems with SCSI interfaces in advanced servers in the early days. With the development of computer technology in recent years, the CPU speed of PCs has entered the GHz era. The hard disks of the IDE interface are also lagging behind, and ATA66 and ATA100 hard disks are successively released. This makes it possible to apply RAID technology to low-end or even personal PCs. RAID is usually implemented by the RAID Controller in the hard disk array or the RAID card in the computer.

With the continuous development of RAID technology, we now have seven basic RAID levels, ranging from RAID 0 to RAID 6. In addition, there are some basic RAID-level combinations, such as RAID 10 (a combination of RAID 0 and RAID 1) and RAID 50 (a combination of RAID 0 and RAID 5. Different RAID levels represent different storage performance, data security, and storage costs. However, we usually use the following RAID forms.

(1) RAID 0
(2) RAID 1
(3) RAID 0 + 1
(4) RAID 3
(5) RAID 5

There are three main factors to choose a RAID level: availability (data redundancy), performance, and cost. If availability is not required, select RAID0 for optimal performance. If availability and performance are important and cost is not a major factor, select RAID 1 based on the number of hard disks. If availability, cost, and performance are equally important, select RAID3 and RAID5 based on the average number of data transmission and hard disks.

Advantages and disadvantages of blade servers: 1. compact structure and high density
The blade server allows more processing capabilities to be inserted into smaller spaces, simplifying cabling, storage, and maintenance. Blade servers are often used for grid computing. The advantage of blade servers is that multiple servers with shared power supplies and other components can be placed in one cabinet, it also allows you to merge storage and network devices and other related resources into a smaller architecture than a cluster composed of common servers.

2. Server Load balancer and Failover
Like most cluster application systems, the blade server has another advantage in managing the blade server, adding load balancing, failover, and other functions. Although this can also be achieved through clusters composed of common servers, the blade server shares much simpler and more compact infrastructure, and is often designed by the vendor for this task, therefore, it is easier to manage Server Load balancer and failover when using blade servers. In case of any hardware fault on the blade server or the chassis itself, the self-diagnosis function will automatically start to take effect, and the indicator display of each server blade can confirm the fault area.

3. Power Consumption and Power Management
Most importantly, blade servers can reduce power consumption and improve power management. The power system is integrated into the blade chassis, reducing the number of independent power supplies required, while reducing the power demand for each server. Because each server blade is streamlined to the maximum extent, there are no other components in the traditional server, such as the keyboard, video card, and other components. Therefore, less power-consuming devices are used. This reduces the overall power consumption. A blade server with 16 server blades consumes much less than 16 servers of full size.

4. Reduce management costs
This is a major advantage over traditional servers. Blade servers are often equipped with an interface for centralized management of each server in the chassis. Server Consolidation and resource centralization simplify server deployment and management.

Hardware configuration management, operation status monitoring, and fault monitoring all implement centralized control, thus reducing the burden on system administrators. The system can promptly notify the administrator of errors by email or other means.

If necessary, the administrator can send out the prepared Disk Images (operating systems and applications ). This feature significantly reduces the time required for system management.

5. Flexibility, scalability, and ease of upgrading

Today, the blade server adopts a special design method to extract and insert the server blade while the system is running normally. You only need to configure it on the administrator interface. New processors, communication components, storage devices, and interconnect devices can be added to the blade, and the blade can be easily loaded into the existing device, which basically has no interference with the functions of the entire system. You can expand or reconfigure the system without stopping your tasks (except for those tasks related to upgrading replaceable server blades ).

The module can be added to the same blade server chassis. The server blades in the chassis are not necessarily the same. For example, depending on the type of the chassis you use, one server blade can be an Intel Xeon architecture, and the other can be an Intel anteng architecture. You can enable one server blade to support Windows, and the other server blade to support Linux.

6. Deployment and scalability

The same is true for adding a new server blade to the system. Once the blade server chassis has been installed, you can add additional servers as long as the additional servers are inserted into the extra bracket while the entire system is running normally.

You only need to add a new server blade to expand the blade as needed. Deploying additional server blades is undoubtedly easier than deploying new independent servers.

Summary

These cannot be learned directly from textbooks. They are in contact with customers in real communication and use. The practice in real life is more vivid and profound in understanding textbooks (thoughts are improved: it is fun to combine learning with life. Life is learning, and learning is life ).

Unfortunately, the Raid of the customer is ready. If you can really understand the Raid blade server, it would be nice. The knowledge of this method is reserved first, when the opportunity comes, let's get started.

Database migration: This was mentioned in the Process of communicating with longge. It takes some time to migrate data temporarily. The merger of old data and new data involves a series of problems ......, This knowledge and Li Xiang (xiang Ge (he studied this some time ago) also communicated). Indeed, database migration is a piece of gold, and we have not completely solved this problem by upgrading the class, for example, the examination system combines multiple servers, Yonghe's new and old databases, and personnel's data integration. At present, we mainly look up data to expand our knowledge. After all, there are very few useful materials on the Internet. We also have our own in-depth research and study; I checked the information on the Internet and saw some large companies dedicated to data migration. It can be seen that this knowledge is really important and the next task is still difficult, when we improve the data in the database class, we can achieve flexible migration, and our database technical capabilities will reach a new level.

Cherish maintenance opportunities and practice to promote learning!