| Selected: |
|---|
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Reset All |
| Selected: |
|---|
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Select: |
| Reset All |
Do you need assistance in RAID planning? Calculates the usable space, space required for data protection, and unused space for different RAID levels. The tool quickly finds out the functional capacity for a given configuration. The device will help you calculate the chances for data loss in a given number of disks in different RAID levels.
RAID is an acronym; stands for Redundant Array of Inexpensive Disks or Redundant Array of Independent Disks. That means RAID is a logical way of putting the several disks together in an array. The idea behind these disks will work together and give the reliability and speed of the expensive disk.
However, the exact speed and reliability you will get from RAID depend upon the type of RAID you are using.
There are several types of RAID. However, the most used ones are given below.
1 - RAID 0: RAID 0 (striping) is all about performance. It takes the number of disks and merges them into a large volume. An individual file can use the speed and capacity of all the disks of an array. That will significantly increase speeds as you are reading and writing from multiple disks at once.
However, it also comes with a downside. RAID 0 is NOT redundant. The loss of any single disk in an array will cause the loss of entire data.
RAID 0 is an excellent choice where you need maximum performance, and the data exists somewhere else. That RAID type is rarely used in server environments. However, you can use it for caching purposes, where you need fast speed, but data loss or reliability does not matter.
2 - RAID 1: RAID 1 (mirroring) is all about fault tolerance and reliability. RAID 1 is competent in a much more complicated configuration. In that RAID type, you have identical disks. These disks have an exact mirror/copy of the data. The main point of RAID 1 is redundancy. If you lose the data of one disk, you still have all your data on the second disk.
Therefore, if in an event where either disk fails. You can replace the broken one within no time. RAID 1 gives you an extra benefit of increased read performance, as you can read off the data from any of the disks in the array. However, the downside is that you will have a slightly higher write latency. As the data needs to be written on both disks in an array, you will only have the available capacity of a single disk while requiring two disks.
3 - RAID 5 and RAID 6: RAID 5 and RAID 6 take the concept of striping plus distributed parity. The simple difference is that RAID 5 requires the use of a minimum of three disks, where RAID 6 requires the use of a minimum of four disks. It uses the RAID 0 concept to strip the data on multiple disks and maintain redundancy by distributing parity information across all the disks. In short, in RAID 5, if you lose one disk, and in RAID 6, you can still make your data and operations functional if you lose two disks.
However, the main drawback is that if you lose a disk in RAID 5 or RAID 6, you will have to sacrifice the performance to make your data operational seriously. Once you replaced the failed disk, it will need a significant array's performance to rebuild the data from the parity information. That rebuild continues to grow more and more as disks data grow.
4 - RAID 10: RAID 10 uses the concept of mirroring and striping. It's a combination of RAID 1 (mirroring) and RAID 0 (striping). In that RAID type, you require a minimum of four disks to increase speed and redundancy. In four disks configuration, you have two or more sets of mirrored disks striped together. That means RAID 10 allows a total of two disks failures, one per mirrored set.
The RAID 50 and RAID 60 are the combinations of RAID 5 and RAID 6 with striping RAID 0 at a higher level.
The process is simple and involves the following steps.