In a previous article, we explained the basics of what is RAID? and what the benefits are of using RAID configurations for your drives.
We thought we’ll continue the theme and talk about the different RAID configurations out there. Not all RAID is equal.
There are generally 6 different RAID configurations. 0,1,5,6 & 10. Each configuration is useful in specific circumstances. Unfortunately, there isn’t an industry standard or committee for RAID configurations which has resulting in companies being able to create their own unique configurations and numbers.
In this article, we will look into RAID 0,1,5,6 and 10.
Where is 2,3,4 and 7 you may ask? These do exist, but they are not that common so we decided to only cover the RAID configurations that you would most likely come across.
RAID 0 – Striping
All drives in the array are used and data is divided into blocks then written across the array. RAID 0 has great I/O performance due to using multiple disks at the same time. The performance can be further enhanced by using multiple controllers – preferably one controller per drive.
Advantages of RAID 0
- Performs great for both read and write operations
- All storage capacity is used, there is no overhead
- Easy to implement
Disadvantages of RAID 0
- RAID 0 isn’t 100% protected from failure. You shouldn’t use it for mission critical operations because if one drive falls, all data in the array would be lost.
Tip: It should be used when working with data that requires high speed read/write such as working graphics and video editing. As mentioned previously, this shouldn’t be used for mission critical systems.
RAID 1 – Mirroring
As the name suggests, RAID 1 duplicates itself and read/writes onto a mirrored set of drives. Controllers can use the mirrored drive for data recovery if the drive fails.
Advantages of RAID 1
- RAID 1 has a solid write speed, similar to a single disk. It also has superb reading speeds.
- Its mirrored drive prevents data from being lost if a drive fails.
Disadvantages of RAID 1
- Duplicating everything isn’t the most efficient way to store data. It only allows for half the capacity of the total drive.
- If your RAID 1 configuration is purely software, if a drive fails you may not be able to recover your data without powering down the system. If this was on a server had multiple users then downtime might be acceptable. To prevent this you can use a hardware controller.
Tip: RAID 1 is great for storing irreplaceable data such as finances due to its availability. Alternatively, it could be used for servers that only use two drives.
RAID 5 – Striping with parity
RAID 5 requires at least 3 drives but you could have up to 16. Each drive is striped and then given a parity data. If a drive fails the parity information can be used to recalculate the information from the data in the other blocks. RAID 5 can be achieved in software but we would advise using a hardware controller here.
Advantages of RAID 5
- RAID 5 has superb data transaction reading speeds and decent writing speeds. The writing speed is reduced due to creating the parity.
- The extra time to write parity is worth it, because if one drive fails then it can use the parity information to recover lost data.
- You can access all your data, even while the controller recovers the data on a new drive.
Disadvantages of RAID 5
- Failing drives have an effect on the throughput.
- Due to the complexity of the technology, it can take quite a bit of time to recover data depending on the controller and size of disk. If another drive fails during this time then all data is lost.
Tip: RAID 5 is very common configuration due to its all round performance and security. It is recommended for applications and file servers with a limited number of drives/
RAID 6 – Striping with double parity.
RAID 6 is similar to RAID 5 but it write double the parity information. RAID 6 requires at least 4 drives, but can still survive even if 2 drives fail. The chances of two drives failing is small, so the chances of three drives failing is even rarer.
Advantages of RAID 6
- RAID 6 also has superb data transaction reading speeds.
- RAID 6 can survive two failed disks and still give access to all your data, even while it is being recovered. Due to this RAID 6 is seen as more secure.
Disadvantages of RAID 6
- Has slower write speeds than RAID 5 due to it creating the second parity.
Tip: RAID 6 is very similar to RAID 5, RAID 6 offers increased security and a good performance. We would recommend using it on application or file servers with a large number of drives.
RAID 10 – RAID 1 & 0
RAID 10 is the combination of RAID 1 and 0. The hybrid combination allows for fast data transfer by stripping the data. It provides redundancy by writing to two discs simultaneously. It is important to note that RAID 10 provides redundancy, but it should not be seen as a backup.
Advantages of RAID 10
- If a drive fails the recovery time is very quick.
Disadvantages of RAID 10
- Mirroring drives isn’t the most efficient way to store data as it halves your capacity. A rebuild after a drive failure also puts a lot of pressure on the remaining drive.
Tip: We would recommend using this configuration on database servers that have high-intensity read/write operations.