RAID disk array

Catalog

Preface

One .RAID Disk array Introduction

Two .RAID The classification of

3、 ... and .RAID Disk array steps

1.RAID 0 To configure

2.RAID 1 To configure

3.RAID 5 To configure

4.RAID 1+0 To configure

---

Preface

disk array (Redundant Arrays of Independent Disks,RAID), Yes " Redundant array of Independent Disks " The meaning of .
Disk array is made up of many cheaper disks , Combine them into a huge disk pack , Use the data provided by individual disks to increase the performance of the whole disk system . Using this technology , Cut data into sections , Stored on each hard disk separately . Raid can also utilize parity checking (Parity Check) The concept of , When any hard disk in the array fails , Data can still be read out , In data reconstruction , Put the data into the new hard disk after calculation .

One .RAID Disk array Introduction

RAID 0 （ Striped storage ）
RAID Divide data in consecutive or byte units , And read and write on multiple disks , Therefore, it has high data transmission rate , But there's no data redundancy .RAID 0 Just simply improve performance , There is no data reliability , And the loss of one hard disk will affect all data .RAID It cannot be applied to occasions with high data security .

RAID 1 （ Mirrored storage ）
Realize data redundancy through disk data mirroring , Generate backup data on a pair of independent disks . When raw data is busy , Data can be read directly from the mirrored copy , therefore RAID 1 Can improve read performance .RAID 1 It's the highest cost per unit in a disk array , But it provides high data security and availability , When a disk fails , The system can automatically switch to the mirror disk for reading and writing , And there's no need to reorganize failed data .

RAID 5（ Stripes and mirrors ）
N（N>=3） The disks form an array , A data generation N-1 A strip , As well as 1 Check data , common N The data is in N Cycle balanced storage on the block disk .N Read and write at the same time , High reading performance , However, due to the problem of verification mechanism , Write performance is relatively low .（N-1）/N Disk utilization is reliable . High reliability , Allow bad 1 Block plate , It doesn't affect all the data .

RAID 1＋0（ Mirror first and then strip ）                  RAID 0＋1（ Strip first and then mirror ）
N The block disks are mirrored in pairs to form RAID 0              Read and write performance and RAID 10 identical
N/2 Disk utilization                                           The safety is lower than RAID 10
N/2 All disks are written simultaneously ,N All disks are read at the same time
High performance , High reliability

Two .RAID The classification of

Level             Number of disks       Redundancy     check       Available disk space
RAID-0        2 Block or more     nothing             nothing         The sum of multiple disks （ Striped storage , Distributed storage , fast ）
RAID-1        2 Block or more     Yes              nothing         2 Sum of disks 1/2（ Copy , Hang one on top of another ）
RAID-5        3 Block above         Yes              Yes         The total capacity of multiple disks (n-1)/n   Allow a broken piece （ There should be a disk verification data 0 and 1）
RAID-6        4 Block above        Yes             Yes         The total capacity of multiple disks (n-2)/n   Allow two pieces to break （ The other is a parity disk ）
RAID-10      4 Block above         Yes             nothing         The total capacity of multiple disks n/2（ Mirror first and then strip ）（ Each group is allowed to break one ）

                Data writing                                 When reading                    
RAID0     Each disk 5m                              from 2 Block disk read         
RAID1     Each disk 10m                            Read performance *1
RAID5     Each disk 5m（+ A calibration disk ）  3 The block plate foundation starts from 2 Fast dish

3、 ... and .RAID Disk array steps

1.RAID 0 To configure

2.RAID 1 To configure

3.RAID 5 To configure

4.RAID 1+0 To configure