Familiarity with RAID and how to use it
What is RAID?
RAID or Redundant Array of Independent Disks means an extension array of Independent disks This term sounds very unfamiliar at first glance, but understanding this technology is easier than you think. The main purpose of RAID is to store the same data on different disks to prevent data loss when the disks are damaged or crashed.
RAID was first introduced in 1987 byDavid Peterson, Randy Katz and Garth Gibson came up. These people considered 6 RAID modes with different characteristics (which were named from 0 to 5), although nowadays the number of RAID modes has increased. had been discussed. IBM filed a patent in 1977 in this field, which was later called RAID 4. In 1983, Digital Equipment released drives with RAID 1 capability. In 1986, IBM filed another patent for what is now known as RAID 5.
How does RAID work?
RAID works by placing data across multiple disks. By allowing input and output of data from all these disks, performance and speed increase. By using multiple disks, error tolerance is increased and data loss is less likely to occur.
RAID arrays are treated as one logical disk by the operating system, and you do not see each disk independently. The size range of data partition units in RAID starts from 512 bytes and reaches several megabytes.
A RAID controller acts as an interface between the operating system and physical disks, and groups of disks as logical units to the operating system. Using a RAID controller improves performance and protects data in the event of a crash.
A RAID controller can be both hardware and It should also be software. In a hardware controller, a physical controller manages the array. This physical controller can be in the form of a PCI Express card. and designed to support SATA and SCSI formats. The physical controller may also be part of the motherboard (motherboard).
The software controller uses system hardware resources to perform its task. They do, but it doesn’t increase performance as much as a hardware controller.
If software controllers are not compatible with the system and hardware controllers are expensive, it is recommended to use RAID based on operating system or driver.
In this type of controller, the controller chip is embedded on the motherboard and All operations are performed by the CPU. So far, this controller works like a software controller; But the difference is that this controller monitors the RAID operation after the system is up. This controller does not have the price of hardware controllers and has higher performance than software controllers.
RAID can be used in various ways. Each of these methods brings its own efficiency and data protection. The following types are more popular and popular:
- RAID 0
- RAID 1
- RAID 5
- RAID 10
In this case, the information stored between the disks You are divided or so-called “stripe“; For example, if we have two disks, half of the information is stored on one disk and the other half on the other disk. Suppose the system wants to store the data “100111” in this mode; The first one, the second zero and the third one will be stored in disk 1 and the rest in disk 2, and finally “101” will be stored in disk 1 and “011” in disk 2.
Advantages of RAID 0
- Increasing read and write speed
- Usability of the entire volume of disks
- Small number of disks required
- Ease of implementation
Disadvantages of RAID 0
- If any of the disks is damaged and its information is lost, practically the entire information is lost. It goes up.
RAID 0 is not at all suitable for sensitive uses, instead it will be useful for tasks such as image and video editing due to its high speed.
In this case, the information on your disks is written the same, and your disks are exactly the same. The information is similar. For example, the data “1001” in this case will be stored as “1001” in all disks.
Advantages of RAID 1
More protection of information
Disadvantages of RAID 1
- Lower speed than RAID 0
- Usable volume is equal to half of the total volume.
- With the damage of one disk, the possibility of Bad Sector and damage in the other disk increases.
RAID 1 can be used in accounting systems and small servers.
This mode of RAID tries to combine the best of RAID 0 and RAID 1 provide you with; Increase speed without increasing the risk of data loss. You must have at least three disks to use this mode. Two disks are written like RAID 0 and data is split between the two. Then the parity bits are written in the third disk. The balance bit is a bit to show whether the number is even or odd, and with the help of this data, if any of the disks is lost, the system can create a new disk with the data of the other two disks, and the information remains undamaged.
Advantages of RAID 5
- Not having the problems of RAID 0 and RAID 1
- High reading speed
- Increasing data protection
Disadvantages of RAID 5
- Minimum number of disks required more than RAID 0 and RAID 1
- Not having the problems of RAID 0 and RAID 1
- Increasing data protection
- Much faster data recovery than RAID 5
- Higher reading and writing speed than RAID 5
- Minimum number of disks required more than RAID 0, RAID 1 and RAID 5
- The number of unusable disks is more than RAID 5 and the significant difference of usable volume in the high number
- Usually by pressing Ctrl+R buttons or Ctrl+A will enter the RAID settings at boot time.
- If RAID is in the BIOS (like most Intel motherboards) it should be entered at boot time Enter BIOS, which is usually done with F2, Ctrl+S and Del.
- If you choose RAID 0 or RAID 5, you will be asked to select the size of each unit, which is better to set to the default mode. You can experimentally increase the RAID efficiency by using different numbers.
- Select Custom (Advanced) installation.
- If you did not see RAID when selecting the Windows installation driver, click Load Driver in Click the bottom section.
RAID 5 is suitable for servers with a limited number of disks because of its speed and stability.
This mode It is actually RAID 1+0. In this case, first the data is divided between a pair of disks like RAID 0 and then this pair of disks is written on another pair of disks like RAID 1. Of course, this operation may be done in the reverse order, which does not result in a change. In this case, at least four disks are needed.
Advantages of RAID 10
Disadvantages of RAID 10
You can visit RAID calculator site, see capacity, speed and fault tolerance in different RAID modes.
|attribute||RAID 0||RAID 1||RAID 5||RAID 10|
|Risk of data loss||High||low||Low||Low|
|Capacity used by RAID||0%||50%||1 disk||50%|
|minimum disk required||2||2||3||4|
|Faultability||none||1 disc||1 disc||1 disc|
Summary of the characteristics of the proposed 4 modes of RAID
How to use RAID?
Note that it is better to use disks with the same volume in RAID to avoid possible problems caused by the volume of the disks not being the same. Also, if your disks contain important files, be sure to make a backup of them.
This method is better due to speed and performance. Compared to the software method, it is suggested. First, you need to see if your motherboard supports RAID or not. To do this, just enter your BIOS (usually the F2 key at the beginning of system boot) and look for the RAID section. If this section is not there, you need to get a RAID adapter and install it on your motherboard.
Each RAID adapter has its own setup program for selecting RAID type and disks. Sometimes these settings are found in the BIOS section. You can get more information in this regard by referring to the motherboard and RAID adapter guide.
To install Windows on RAID (Windows Vista and above), pay attention to the following points:
Using software RAID (RAID provided in Windows)
Windows XP, Vista (Vista), seven (7), eight (8) and ten (10), all offer software RAID as an alternative to hardware mode and are easy to use.
Follow the procedure below to change RAID Use Windows:
1. Right click on Computer and select Manage.
2. Click on Disk Management which is under Storage.
3. Right-click on the large drive icon on the left to see operators that affect the entire disk. Only options that are possible on your disk are enabled. Sometimes, if your disk is not dynamic (Dynamic), a message will be displayed that by following the instructions, your disk will automatically become a dynamic disk.
4. Right-click on dynamic disks and you will see an option to copy all data similar to RAID 1 (this feature is only available in Windows 7 Professional and Ultimate editions, Windows 2000 Ultimate and Windows Server).
Using RAID in Windows 10
1. First, write Storage Spaces from the Windows search bar next to Start.
2. In the opened window, select Create a new pool and storage space. Windows will select all drives in the next window.
3. In the Resiliency section, specify the type of RAID. The two-way mirror and three-way mirror options are similar to RAID 1, and Parity is similar to RAID 5. .
4. If necessary, set the drive size. By default, the correct value is selected.
5. Select Create storage space.
6. Wait until the process is complete.
7. To manage the array, write Manage Storage Spaces in the search bar.
The future of RAID
Although RAID is still used, many analysts believe that with the advent of alternative technologies such as erasure coding that offer better data protection (albeit at a cost). higher), the RAID is declining. These technologies are being developed with a focus on solving RAID problems; Problems such as data protection and increasing the probability of errors with the increase in the number of disks.
Emergence of solid state drives (SSD) have reduced the need for RAID. SSDs have no moving parts and therefore do not have the instability problems of hard disks. SSD arrays usually use techniques such as Wear Leveling instead of RAID to protect data. Hyperscale computing also eliminates the need for RAID by using servers instead of drives.
However, RAID is still considered an integral part of data storage, and major technology companies continue to offer RAID products. IBM has released its own RAID product with Spectrum Virtualize V7.6, promising to improve RAID performance. The latest version of Intel Rapid Storage technology supports RAID 0, RAID 1, RAID 5 and RAID 10. NetApp ONTAP management software uses RAID to protect data against the simultaneous loss of up to 3 drives. Dell EMC’s integrated platform also supports RAID 0, RAID 1, RAID 5 and RAID 6.
What do you think about this technology? What do you know about the strengths and weaknesses of this technology and what technologies can replace RAID in the coming years? Share your views with us.