Anyone who works with data storage and information security certainly must have come across the term RAID along the way. However, although some even use their concepts, not everyone knows exactly what RAID is and what types of RAID exist.
Understanding how this mechanism works is important to discover how this system is able to provide more security and performance to your applications.
We will explain in detail to you not only the concept but also what are the different types of RAID that exist. The idea is quite simple and after understanding what each item means it becomes easier to understand the logic of its operation.
What is RAID?
RAID is an acronym for Redundant Array of Inexpensive Disks. In direct translation, the term means something like Redundant Matrix of Independent Discs.
Got even more lost? Calm. Translating this information: it is the combination of several hard drives, the popular HDs so that they form a single logical unit. In this way, the same data that is stored on one disk is available on another.
In other words, when several hard drives work as one, we have a RAID. And why is this useful? In the event of a disk failure, the others will continue to function, preventing information from being lost or interrupting access to data. Thus, we can say that RAID provides more security and more performance. It can act as a redundancy mechanism, for example.
What types of RAID are there?
There are several types of RAID that can be performed with storage disks. Each has a different purpose. Some are more focused on performance while others are on security. There is no “right” choice to make, as each occasion and circumstance will call for a different decision. In case if you want to know in detail about RAID storage read more In general, we have the following RAIDs:
Also known as “fractionation”. In this case, the available data is divided into small segments and distributed across the disks. There is no redundancy here and therefore there is no fault tolerance.
Which means that in case of problems, data may be lost. On the other hand, this feature improves the computer’s performance because it will access data faster.
- Very quick to access information
- It has low cost in memory expansion
- Has no mirror
- No data parity
- If some of the HD sectors fail, the divided file may become unrecoverable
In this model, the basis is the mirroring of one disk on another. In other words, it is as if there were a copy of disk A on disk B and vice versa. In addition to the advantage of being more secure compared to RAID 0, there is virtually no loss of performance. Because of this, this format is widely used on servers.
- Data security (regarding HD defects)
- If any sector fails, you can recover by copying files from another HD
- Has mirroring
- Parity is not used
- Writing takes longer
- Higher cost compared to RAID 0
Little used, RAID 2 practically fell out of use because the new hard drives are already out of the factory with mechanisms similar to it and that prevent certain failures. What this system does is to detect failures in hard drives and, therefore, it starts to work for error checking. In short, all disks are constantly monitored by this mechanism.
- Uses ECC (Error Correction Code) technology, which reduces error rates to almost 0 even with power failure
- Today’s hard drives already have this technology natively
- Depending on the configuration, space may be wasted
Here all information is divided on the matrix disks. The exception is for one of them, who becomes responsible for storing parity data. Its greatest advantages are the possibility of transferring large volumes of data and reliability in protecting information.
- Reads and writes very fast
- Has error control
- Its assembly via software is more complex
RAID 4 is similar to RAID 3, but here the data is divided between the disks. The great advantage of this version is the possibility of reconstructing the data through the parity mechanism. It is the most suitable option when it comes to large files, as it is the one that best guarantees the integrity of the information.
- Very fast reading rate
- You can increase the area of physical disks
- Recording rate is slower
- In comparison to RAID 1, its reconstruction in case of disk failure is more complex
- Old technology compared to newer systems
RAID 5 is the natural evolution of versions 2, 3 and 4. Here, the space equivalent to an entire disk is reserved for storing parity information.
The system here is more complex and the parity is stored alternately on several disks. So, if any of them has a problem, just start a process called rebuild to retrieve all the information.
- Faster to spot errors
- Quick reading
- Slow writing
- More complex disk control system
RAID 6 (Dual Parity)
It is basically the same case with RAID 5, but with twice the parity bits. In this case, even if two hard drives fail at the same time, the data will not be lost. It is one of the safest alternatives available.
- Possibility to fail 2 Hds at the same time without data loss
- It needs at least 3 Hds to be implemented because of parity
- Your writing is slower
- Data control is also complex
Finally, we have RAID 10, a system that borrows characteristics from RAIDs 0 and 1. This system can only be used with more than 4 disks and always in an even number. In this case, half of the disks store data and half make copies of it. It is the safest of all.
- Security against data loss
- One or two hard drives may fail at the same time (depending on which failure)
- High expansion cost
- Drivers must be in speed sync to increase performance