1. What is RAID?
RAID is the abbreviation of English Redundant Array of Independent Disks , and the Chinese abbreviation is Redundant Array of Independent Disks . Simply put, RAID is a technology that combines multiple independent hard disks (physical hard disks) in different ways to form a hard disk group (logical hard disk), thereby providing higher storage performance and data backup than a single hard disk . In fact, it uses multiple independent disks to form a large disk system, thereby achieving better storage performance and higher reliability than a single disk.
A secondary hard drive protects the server and can also increase storage capacity, speed up access, etc.
2. RAID terminology:
Striping: refers to dividing data across multiple drives. Striped RAID arrays are typically used to combine maximum capacity into a single volume.
Mirroring: refers to copying data to multiple disks. Depending on the array's RAID level, mirrored RAID arrays generally ensure data loss if (at least one) disk in the array fails.
Fault tolerance: refers to the ability of a RAID array to continue working when a disk fails (that is, users can still use the data stored in the array).
3. RAID classification
After continuous development, RAID technology now has eight basic RAID levels from RAID 0 to 7. In addition, there are some combinations of basic RAID levels, such as RAID 10 (a combination of RAID 0 and RAID 1), RAID 50 (a combination of RAID 0 and RAID 5), etc. Different RAID levels represent different storage performance, data security and storage costs
RAID can be divided into different types of RAID implementations based on the number and organization of hard drives.
Common ones are: RAID 0, RAID 1, RAID 5, RAID 01 (0+1), RAID10 (1+0).
Less common ones are: RAID 2, RAID 3, RAID 6, RAID 4, RAID 7, RAID 50, RAID 53.
4. Common introduction to RAID
RAID 0
RAID 0 is the fastest RAID mode
RAID 0 requires at least two disks and stripes data to each disk
The available capacity of all disks is combined into one logical volume on the computer
If one disk fails, the data on all disks will be inaccessible because the data is striped across each disk.
Advantages: striping, parallelism, full utilization of bandwidth, high read and write efficiency, high storage space utilization (100%), and low cost.
Disadvantages: no backup, poor reliability, no redundancy, no verification, in principle it is not a RAID structure.
Summary: RAID 0 is ideal for applications requiring maximum speed and capacity . RAID 0 arrays are more suitable for use in frequent file processing and should not be used as the only storage backup solution. They are also not suitable for use in mission-critical systems and cannot be used. In situations with high data security requirements.
RAID 1
RAID 1 is a secure RAID mode
RAID 1 requires at least two disks and a logical volume will be installed on the computer.
The available capacity after the two disks are combined is limited to the maximum capacity of the smallest disk.
If one disk fails, data can be fetched immediately from the second disk
Even if one disk fails, no data is lost
Data redundancy is achieved through disk data mirroring, at least 2 disks
Advantages: Provides redundancy, high data security and availability, higher reading efficiency than a single hard drive, and supports "hot swap"
Disadvantages: low performance, high cost, lower writing efficiency than a single hard drive,
Summary: Focus more on security than speed
RAID 5
RAID 5 combines the striping technology of RAID 0 and array data redundancy technology
RAID 5 至少需要三个磁盘
RAID5模式中，不再需要用单独的磁盘写校验码了。它把校验码信息分布到各个磁盘上。
一旦某一块磁盘损坏了，就可以用剩下的数据和对应的奇偶校验码信息去恢复损坏的数据。
单个磁盘出现故障时，数据不会丢失，但如果数据尚未重建到更换驱动器上，而此时又有一个磁盘出现故障，那么阵列中的所有数据都将丢失。
优点：读写性能高，有校验机制，空间利用率高。
缺点：组成 RAID 5 的磁盘越多，安全性能越差，容易丢失数据连续两块硬盘损坏，数据就找不回来了。
总结：RAID 5 适用于 归档，且 适合追求性能并要求持续访问数据（如：视频编辑）
RAID 01（0+1）和RAID10（1+0）
RAID 0 + 1 是一种 安全 的 RAID 模式，由条带集的镜像组成，条带化后镜像 合并了其他级别（尤其是 RAID 1 和 RAID 0）特点
RAID 0 + 1 的 磁盘数 为 四的倍数
在 RAID 0 + 1 中，五个磁盘 的 第五个磁盘 将成为 备用磁盘 或 空置磁盘
在 RAID 0 + 1 中，最多允许两个磁盘出现故障 而 不会丢失数据，但故障磁盘 必须属于同一 RAID 0 队列，如图也就是当磁盘 1 和磁盘 2 出现故障时，数据会保存到磁盘 3 和磁盘 4。
优点：是同时拥有RAID 0的超凡速度和RAID 1的数据高可靠性
缺点：成本高，CPU占用率高，而且磁盘的利用率比较低
RAID10（1+0）
RAID 10（也称为 RAID 1 + 0）是合并了其他级别（尤其是 RAID 1 和 RAID 0）特点的另一种 RAID 级别
RAID 10 是一种 镜像集条带，意思是 数据在两个镜像阵列间分条，镜像后条带化，条带化 在阵列之间发生，而 镜像 是 在相同的阵列中出现，两种技术的组合加快了重建的速度
RAID 10 的 磁盘数 为 四的倍数
在 RAID 10 中，五个磁盘 的 第五个磁盘 将成为 备用磁盘 或 空置磁盘
在 RAID 10 中，每个镜像 对中 可以有一个磁盘出现故障 而 不丢失数据
故障磁盘 所在阵列的磁盘会成为整个阵列中的弱点
如果 镜像 对中的 另一个磁盘也发生故障，则 会丢失整个阵列
RAID 10 使用 RAID 0 条带技术 来 提供良好的速度，但设备的 可用容量会减少一半
RAID 10和RAID 01
共性：都是RAID 0和RAID 1的结合方案，继承了RAID 0和RAID 1的优点，在数据安全性和数据读写效率上都有保障
异性：条带化和镜像先后顺序的区别
RAID 01允许坏多块磁盘，但只能在坏在同一个RAID 0中，不允许两个RAID 0都有坏盘。
RAID 10允许坏多个盘，只要不是一对磁盘坏就可以。
参考帖子
知乎
https://zhuanlan.zhihu.com/p/129295334
https://zhuanlan.zhihu.com/p/119452913
CSDN