Redundant Array of Independent Disks (RAID)

What is RAID (Redundant Array of Independent Disks)?

RAID or Redundant Array of Independent Disks refers to a collection of hard drives set up and connected together to enhance the performance of disk storage and offer additional protection to the stored data, offering fault tolerance.

Understanding RAID (Redundant Array of Independent Disks)

RAID is a data storage virtualization technology where several physical disk drives are combined to form a logical unit or units for data redundancy and performance enhancements.

It basically stores the same data on different disks and places. These drives work in parallel. This prevents data loss in the event of a drive failure.

The connected disks are called the array members and are combined using different RAID levels.

The RAID systems use different interfaces such as:

• SATA
• SCSI
• IDE
• FC

Input/output operations are overlapped in a RAID system in a balanced way and the array appears as a single drive to the operating system.

RAID Levels

Typically, the RAID levels are categorized as standard RAID levels, non-standard RAID levels, and hybrid or nested RAID levels. There are RAID 0, 1, 2, 3, 4, 5, 6, 53, 0/1, and more, though some of them are not used today.

Here are the standard and commonly used RAID levels.

RAID 0: Striping

Also known as a striped volume or striped set, RAID 0 can be done with two drives only. This specific level is characterized as follows:

• The drives are merged together to form one large volume to store data.
• The data is split into different blocks before being stored.
• The data is written or read sequentially or simultaneously on the disks which increases the performance.
• It is fast and offers more efficiency.
• Storage space will be limited to double the size of the smallest disk if different drive sizes are used.
• It is affordable and allows for using multiple controllers to enhance its performance.
• It is easy to set up and does not include fault tolerance, redundancy, or parity.

Raid 0 should be used when:

• Performance is a priority.
• Reliability is not important.
• Losing data is an option.
• Non-critical storage is needed.

RAID 1: Mirroring

RAID 1 also needs at least two drives to store the same data on them and ensure redundancy. It is commonly used to create a mirrored pair of one disk on the other, which is why it is also known as mirroring.

This specific RAID level can be characterized as follows:

• The primary focus is on providing redundancy rather than performance and speed.
• Since it creates a replica, the chances of data loss and downtime are eliminated in the event of a drive failure.
• The array volume is as large as the smallest disk in this setup.
• It is functional as long as one of the two drives is operational.
• It enhances reliability and read performance because a request can be taken by either one of the two drives.
• The write performance is however the same because the other disk in the setup is equal to the slowest disk.

Raid 1 should be used when:

• You need mission-critical storage.
• The risk of data loss needs to be kept to a minimum.
• High reliability is required.
• You use small servers.
• You are looking for an easy and simple setup.

Raid 5: Striping with Parity

RAID 5 is one of the most commonly used RAID levels and is also considered to be the most secure of all RAID implementations. It can be characterized as follows:

• It combines parity and striping.
• It offers a reliable and quick setup.
• It provides both storage usability and performance efficiency.
• It needs at least three hard drives and can support up to sixteen of them.
• The data is split into strips before being stored across the disks in the array.
• It offers high performance rates.
• It allows faster read data transactions by doing it simultaneously on multiple drives in the array.
• It ensures an even distribution of parity bits on all disks after saving every data sequence.
• It offers redundancy through parity bits rather than mirroring.

Raid 5 should be used when:

• You use file and application servers.
• You need high efficiency.
• You need optimal storage.
• You want a cost-effective and probably the best solution.
• You want constant data access.
• You need to install an operating system within the array.

Raid 6: Striping with Double Parity

RAID 6 is quite similar to RAID 5, but with an additional parity feature. This is why it is also called the double-parity RAID and is characterized as follows:

• It needs a minimum of four hard drives.
• It utilizes block-level striping to spread data across.
• In each data block there are two parity blocks stored.
• Its features allow the failure of two drives before the data is lost.
• The performance depends on the number of drives used and how the array is used.
• The double parity feature makes its operation a bit slower in comparison.

Raid 6 should be used when:

• You need a reliable solution for mission-critical applications.
• You cannot afford data loss.
• Data management is a priority in banking, healthcare, and defense sectors.

Raid 10: Mirroring with Striping

Also known as RAID 1+0, RAID 10 is a hybrid or nested type of RAID configuration where there are two different levels combined into one. It can be characterized as follows:

• It is a combination of RAID 1 mirroring and RAID 0 striping.
• The data is written on two or more drives to offer redundancy by using logical mirroring.
• Blocks of data are distributed across several drives by using block-level striping.
• It offers enhanced performance.
• It allows faster read and write speeds since the data is accessed simultaneously from many disks.

Raid 10 should be used when:

• You need to store high volumes of data such as on web hosting servers, email servers, and databases.
• You need faster read and write operations.
• You want high fault tolerance.

Non-standard RAID:

These are the RAID configurations created by different companies and open-source projects according to their needs. Hence, these are non-standard RAID implementations. Some of these implementations are:

• Drive Extender
• Linux MD RAID 10
• RAID-DP
• RAID-Z
• Declustered RAID

Nested (Hybrid) RAID:

These RAID configurations are designed by combining two or more standard RAID implementations to offer higher redundancy and performance.

These RAID implementations are named according to the RAID levels combined. There are usually two digits in the order of the layering scheme.

Some of the popular hybrid or nested RAID levels, including RAID 10 mentioned above, are:

• RAID 01 – It combines striping and mirroring and is also known as ‘mirror of stripes.
• RAID 03 – It combines byte-level striping and dedicated parity.
• RAID 10 – It combines disk mirroring and straight block-level striping.
• RAID 50 – It combines distributed parity and straight block-level striping.
• RAID 60 – It combines dual parity and straight block-level striping.
• RAID 100 – It is a stripe of RAID 10s.

Read Also:  What is Run Length Limited (RLL)? (Explained)

Do You Need RAID on Your PC?

Yes, you will need RAID on your computer because it is extremely useful for protecting your data. You should use it all the more if availability and uptime are two vital things for you.

RAID is also useful when you face IO issues with your hard disk resulting in long waits for it to complete a given task. You will be able to read and write from other drives instead of adding one to the output.

Moreover, if you have hardware RAID, the card will have additional memory to use as cache. This will also reduce the workload on the physical hardware of the computer to increase its overall performance.

However, you may not opt for using RAID if you are okay with data loss or several hours of downtime while restoring your site from the backups.

There are different types of RAID configurations that you can use such as:

• RAID 0 – It is good to use if performance is critical but data is not.
• RAID 1 – It is good to use if you want added data redundancy or read speeds inexpensively.
• RAID 5/6 – These are good to use if you are using web servers in a read- and write-heavy environment with excessively large storage arrays.
• RAID 10 – It is good to use as an all-around solution that will provide high performance with further data redundancy and read and write speeds.

How to Set Up a RAID?

The process to setup RAID is pretty simple and will vary a bit depending on the specific type of RAID you want to set up.

Typically, you can set up RAID 5 for redundancy and efficiency or the nested RAID 10 system for sheer performance on your Windows 10 by using the System Settings.

To setup RAID 5, you will need three drives for parity and striping and to follow these steps:

• Click on the Start button
• Open the Settings tab
• Select the System tab
• Select the Storage tab
• Go to the More Storage Settings heading
• Select Manage Storage Spaces
• Select the “Create a new pool and storage space” option
• Click on Yes when prompted to accept changes to your system
• Choose the drives you want to pool
• Click on Create pool
• Complete the name and drive letter fields
• Go to the Resiliency heading
• Click on the resiliency type dropdown
• Select Parity
• Go to the Size heading
• Enter the desired size and capacity settings
• Click the Create Storage Space button

To set up a nested RAID 10 system, you will need four drives for mirrored striping and to follow these steps:

• Go to the Start button
• Open Settings
• Select System
• Select Storage
• Go to the More Storage Settings heading
• Select Manage Storage Spaces
• Choose “Create a new pool and storage space”
• Click on Yes to accept changes to your computer when prompted
• Go to the drive selection menu
• Create two different pools
• Choose a pair of drives
• Click on Create pool to setup the first mirror
• Complete the name and drive letter fields
• Go to the resiliency heading
• Click on the Resiliency type dropdown
• Choose Two-way mirror
• Go to the Size heading
• Enter the capacity and size settings wanted
• Click the Create Storage Space button to activate the first mirrored drive pair

To create the second mirrored drive pair, repeat the same steps. You may create as many as you want by following the same process.

Now, press Windows and X at the same time to open Windows Disk Management and follow these steps:

• Look at the bottom for the two mirrored pairs created
• Right-click on the first pair
• Select Delete Volume to erase the file systems
• Repeat the same for the second drive and for others
• Right-click one of the drive pairs you need in the array
• Choose New Striped Volume
• Select the other mirrored pairs in the new window for the drive you want in the array
• Click on Next
• Enter the desired settings
• Click on Next
• Review the settings
• Click Finish

Read Also:  What is Address Register? Uses, Size & More

To complete the process, click on the prompt, if any, and wait until Windows completes formatting the new volume.

Based on where the RAID process takes place, the implementation process may also vary.

For example, for a hardware-based RAID installation, you can use a RAID controller card and insert it into a fast PCI Express slot on your motherboard to connect it to the drives.

For a software-based RAID setup, you will need to connect the drives to the computer system directly instead of using a RAID controller.

The disks, in this case, need to be managed by a utility software on the operating system.

And for a firmware- or driver-based RAID, it hardly needs any separate setup since it is stored on the motherboard directly.

All of its operations are carried out by the CPU of the computer and not by a dedicated processor.

Advantages

• Prevents unintended data loss
• Distributes wear and tear across drives
• Allows storing large data volumes
• High performance
• Reliable since work does not stop even if one drive fails
• Lower write latency
• Proper utilization of disks
• Clear and better communication
• Better risk management

Disadvantages

• No prior warning of drive failure
• Takes long time to rebuild large arrays
• Part of the drive cannot be read if the second drive fails
• Surplus information may create confusion
• Chances of correlated failures
• Unrecoverable read errors while rebuilding
• Write-cache reliability concern
• Atomicity

Questions & Answers:

Is RAID a Backup?

Ideally, you should not consider RAID as a backup, even though it offers additional data protection.

Therefore, make sure you have a proper backup of your data, even if you use RAID. It will save you from any catastrophic data loss.

Which RAID Is Most Commonly Used?

Usually, RAID 5 is the most commonly used RAID configuration. The main reason for this is its data striping utilization.

It involves separating data into segments and storing them on different disk drives in the array.

Where Is RAID Commonly Used?

RAID is most commonly used in servers but is also used in high performance computers, including desktop and laptop computers. Floor-standing RAID units are also used in large Storage Area Networks (SANs).

Which RAID Is Best for Storage?

Different levels of RAID will offer different benefits, but if you are looking for high performance from your storage systems, you should opt for RAID 0.

This offers fast read and write speeds, along with a lot of raw storage capacity.

The best thing is that RAID 0 will not provide any data redundancy, which is commonly associated with RAID, thereby enhancing its performance level.

How Many Drives Are Required for RAID?

You will need a minimum of three drives to set up RAID on your computer. This helps RAID sustain the loss in the event of one drive failing.

In such a situation, the data from the defunct drive will be reconstructed on the remaining drives using parity stripes. However, for some RAID levels, you will need four drives, such as in RAID 10.

Is RAID Used in Laptops?

Usually, RAID is used on servers. However, it has also been used recently in high-performance laptop computers as well as desktop PCs for more efficient distribution of data on the disks.

Do Motherboards Have RAID?

Yes, most of the motherboards today have RAID built-in, but that does not mean every motherboard will support the same RAID configuration.

Typically, RAID uses a specific controller and two or more physical drives. These are typically connected to motherboards that do not come with RAID circuits.

Does RAID Use CPU?

Yes, hardware RAID will typically use a general-purpose CPU, which is commonly a Power or an Advanced RISC processor.

It may even use a custom Application Specific Integrated Circuit (ASIC) to handle the latency introduced in order to prevent the drop in the output.

Ideally, the latency will vary depending on the processing capacity of the system.

Conclusion

RAID is a very useful and efficient way to store and protect your data by storing the same data on separate hard disks and SSDs.

Since there are multiple disks containing the data, you will not suffer any data loss if one particular drive on your computer fails.

There are many levels of it but all do not provide redundancy.