Shulou(Shulou.com)06/03 Report--

1 、 RAID 0

RAID 0 is the earliest RAID schema, that is, Data Stripping data striping technology. RAID 0 is the simplest form of building a disk array, which only requires more than 2 hard disks, has a low cost, and can improve the performance and throughput of the whole disk. RAID 0 does not provide redundancy or error repair capabilities, but the implementation cost is the lowest.

RAID 0 schematic diagram

The simplest way to implement RAID 0 is to create a large volume set by cascading N blocks of the same hard disk in the form of hardware through an intelligent disk controller or using disk drivers in the operating system. In use, computer data is written to each hard disk in turn, and its biggest advantage is that it can multiply the capacity of the hard disk. If the hard disk using three 80GB is built into RAID 0 mode, then the disk capacity will be 240GB. In terms of speed, the speed of each individual hard disk is exactly the same. The biggest disadvantage is that if any hard disk fails, the whole system will be destroyed, and the reliability is only 1max N of a single hard disk.

To solve this problem, another mode of RAID 0 has emerged. That is, select a reasonable band on N hard drives to create a band set. The principle is that the data written in the original sequence is distributed to all four hard drives to read and write at the same time. The parallel operation of four hard drives increases the disk read and write speed by four times at the same time.

When creating a band set, it is very important to choose the size of the band reasonably. If the band area is too large, the band space on one disk may be able to meet most of the Imax O operations, so that the reading and writing of data is still limited to a small number of one or two hard disks, which can not give full play to the advantages of parallel operation. On the other hand, if the band area is too small, any I _ swap O instruction may cause a large number of read and write operations and occupy too much controller bus bandwidth. Therefore, when creating a band set, we should carefully choose the size of the band according to the needs of practical application.

Although the band set can evenly distribute the data to all disks for reading and writing. But if we connect all the hard drives to one controller, it may bring potential harm. This is because it is easy to overload the controller or bus when we read and write frequently. To avoid the above problems, it is recommended that users can use multiple disk controllers. The best solution is to equip each hard disk with a special disk controller.

Although RAID 0 can provide more space and better performance, the whole system is very unreliable and cannot be remedied if there is a failure. Therefore, RAID 0 is generally used only in situations where data security is not required. [1]

2 、 RAID 1

RA

RAID 1 schematic diagram

ID 1 is called disk mirroring. the principle is to mirror the data from one disk to another, that is, when the data is written to one disk, it will generate an image file on another idle disk, so as to maximize the reliability and repairability of the system without affecting performance, as long as at least one disk in any pair of mirror disks in the system can be used. Even when half of the hard drives have problems, the system can run normally. When a hard disk fails, the system will ignore the hard disk and use the remaining mirror disk to read and write data, with good disk redundancy. Although this is absolutely safe for data, the cost will also increase significantly, with a disk utilization of 50%. For four hard drives with 80GB capacity, the available disk space is only 160GB. In addition, the RAID system with hard disk failure is no longer reliable, so the damaged hard disk should be replaced in time, otherwise the remaining mirror disk will also have a problem, then the whole system will crash. After the replacement of the new disk, the original data will take a long time to synchronize the mirror, and the external access to the data will not be affected, but the performance of the whole system will decline. Therefore, RAID 1 is often used in situations where critical and important data is saved.

RAID 1 mainly implements disk mirroring through secondary reads and writes, so the load on the disk controller is also quite heavy, especially in environments that need to write data frequently. In order to avoid performance bottlenecks, it is necessary to use multiple disk controllers.

3 、 RAID0+1

From RA

RAID0+1 schematic diagram

On the name ID0 + 1, we can see that it is a combination of RAID0 and RAID1. When we use RAID 1 alone, there will also be problems like using RAID 0 alone, that is, we can only write data to one disk at a time and can't make full use of all the resources. To solve this problem, we can establish a band set in the disk mirror. Because this configuration combines the advantages of band set and mirroring, it is called RAID 0room1. With the combination of RAID0 and RAID1 technology, in addition to distributing data on multiple disks, each disk has its own physical mirror disk, which provides full redundancy, allows one disk failure without affecting data availability, and has fast read / write capabilities. RAID0+1 needs to establish a band set of at least 4 hard drives in the disk mirror.

4. RAID: LSI MegaRAID, Nytro and Syncro

MegaRAID, Nytro, and Syncro are all LSI solutions for RAID and are constantly creating updates.

The main positioning of LSI MegaRAID is to protect data and provide high-level protection for data through high-performance and reliable RAID controller functions. LSI MegaRAID has a good reputation in the industry.

The main positioning of LSI Nytro is data acceleration, which makes full use of today's popular flash memory technology to greatly improve the speed of data. LSI Nytro includes three series: LSI Nytro WarpDrive accelerator card, LSI Nytro XD application acceleration storage solution and LSI Nytro MegaRAID application accelerator card. Nytro MegaRAID is mainly used in DAS environment, Nytro WarpDrive accelerator card is mainly used in SAN and NAS environment. Nytro XD solution consists of Nytro WarpDrive accelerator card and Nytro XD intelligent cache software.

The positioning of LSI Syncro is mainly used for data sharing to improve the availability and scalability of the system and reduce the cost.

LSI provides basic reliability assurance through MegaRAID, acceleration through Nytro, and capacity bottlenecks through Syncro, allowing low-cost storage solutions to scale on a large scale and further improve reliability.

5. RAID2: with hamming code check

Conceptually, RAID 2 is similar to RAID 3 in that both distribute data in bits or bytes across different hard drives. However, RAID 2 uses certain coding techniques to provide error checking and recovery. This coding technology requires multiple disks to store, check and recover information, which makes the implementation of RAID 2 technology more complex. As a result, it is rarely used in a business environment. On each disk on the left side of the image below are the bits of the data, and the hamming check code calculated by different bits of one data can be saved on another set of disks. Due to the characteristics of hamming code, it can correct errors in the case of data errors to ensure that the output is correct. Its data transfer rate is quite high. If you want to achieve an ideal speed, you'd better improve the hard disk that stores the parity check code ECC code. For the design of the controller, it is simpler than RAID3,4 or 5. There is no free lunch, and it is the same here. If you want to use hamming code, you must pay the price of data redundancy. The rate of output data is equal to that of the slowest in the drive group.

6. RAID3: parallel transmission with parity code

This kind of check code is different from RAID2 in that it can only check errors but not correct them. It processes one band at a time when accessing data, which improves read and write speed. The check code is generated when the data is written and saved on another disk. When it needs to be implemented, the user must have more than three drives, and the write rate and readout rate are very high, because there are fewer parity bits, so the computing time is relatively less. It is very difficult to realize the RAID control with software, and the realization of the controller is not very easy. It is mainly used in graphics (including animation) and other situations that require high throughput. Unlike RAID 2 original raid 3, a single disk is used to store parity information. If a disk fails, parity disks and other data disks can regenerate data. If the parity disk fails, the use of the data will not be affected. RAID 3 can provide a good transfer rate for a large amount of continuous data, but for random data, parity disks will become the bottleneck of write operations.

7. RAID4: independent disk structure with parity code

RAID4 is very similar to RAID3, except that it accesses data by block, that is, by disk, one disk at a time. As you can see in the figure, RAID3 is one bar at a time, while RAID4 is one bar at a time. Its features are similar to those of RAID3, but when recovering from failures, it is much more difficult than RAID3, the controller design is much more difficult, and the efficiency of accessing data is not very good.

8. RAID5: independent disk structure of distributed parity

As can be seen from its schematic diagram, its parity code exists on all disks, where p0 represents the parity value of band 0, and other meanings are the same. The reading efficiency of RAID5 is very high, the writing efficiency is average, and the block collective access efficiency is good. Because the parity codes are on different disks, the reliability is improved. However, it can not solve the parallelism of data transmission, and the design of the controller is also very difficult. The important difference between RAID 3 and RAID 5 is that every data transfer in RAID 3 involves all array disks. For RAID 5, most data transfers operate only on one disk and can be operated in parallel. There is a "write loss" in RAID 5, that is, each write operation produces four actual read / write operations, including two reads of old data and parity information, and two writes of new data and parity information.

9. RAID6: independent disk structure with two kinds of parity codes for distributed storage

The name is long, but if you look at the picture, you will immediately understand why. Please note that p0 represents the parity value of band 0, while pA represents the parity value of block A. It is an extension of RAID5 and is mainly used in situations that require that the data must not go wrong. Of course, due to the introduction of the second parity value, two disks are needed. At the same time, the design of the controller becomes very complex, and the writing speed is not good. It takes a lot of time to calculate the parity value and verify the correctness of the data, resulting in unnecessary load. I don't think anyone can afford this kind of thing except the army.

10. RAID7: optimized disk structure for high-speed data transfer

All the RAID7's Icano transfers are synchronized and can be controlled separately, which improves the parallelism of the system and the speed at which the system accesses data; each disk has cache memory, and the real-time operating system can use any real-time operating chip to meet the needs of different real-time systems. Allows management and monitoring using the SNMP protocol, and an independent transport channel can be specified for the check area to improve efficiency. Multiple hosts can be connected, because with the addition of cache, when multiple users access the system, the access time is close to zero. Due to the parallel structure, the efficiency of data access is greatly improved. It should be noted that it introduces a cache, which has both advantages and disadvantages, because once the system is powered off, all data in the cache will be lost, so you need to work with UPS. Of course, things that are so fast are also very expensive.

11. RAID10: high reliability and efficient disk structure

This structure is nothing more than a band structure plus a mirror structure, because the two structures have their own advantages and disadvantages, so they can complement each other to achieve the purpose of both high efficiency and high speed. We can combine the advantages and disadvantages of the two structures to understand the new structure. The price of this new structure is high and its expandability is not good. It is mainly used in databases with small data capacity but requiring speed and error control.

12. RAID53: efficient data transfer disk structure

The later structure is a repetition and reuse of the previous structure, this structure is the unity of RAID3 and band structure, so it is faster, but also has fault-tolerant function. But the price is very high and it is not easy to achieve. This is because all data must go through both striping and bit-by-bit storage, and it is not easy to require these disks to synchronize in consideration of efficiency.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.