Create RAID Group
Overview
This function creates a RAID group.
A RAID group is a group of drives that configure a RAID level.
Features and Required Number of Drives for Each RAID Level
The following table shows the features and the required number of drives for each RAID level.
Actual RAID group configurations (or the number of available drives) vary depending on the maximum number of drives that can be installed in the storage system.
RAID level (*1) |
Feature |
Required number of drives (*2) |
---|---|---|
High Performance (RAID1+0) |
The high I/O performance of RAID0 (striping) is combined with the reliability of RAID1 (mirroring). |
2D+2M - 16D+16M |
High Capacity (RAID5) |
Data divided into units of blocks and parity information that is created from the data are allocated across multiple drives to allow data redundancy. |
2D+1P - 15D+1P |
High Reliability (RAID6) |
The use of double parity allows the full recovery of lost data even in the event that two of the drives fail. |
3D+2P - 14D+2P |
High Reliability (RAID6-FR) |
Configure a single RAID group with multiple RAID redundant sets and reserved areas equivalent to a hot spare. Distributing data in RAID groups allows high-speed rebuilding when the first drive fails. A recovery can be performed with up to two drive failures, however when the second drive fails, the rebuild is operated at a normal speed. Several restrictions apply to the RAID groups and volumes of this RAID level. Refer to "Restrictions for RAID6-FR" for details. |
(3D+2P)x2+1HS (4D+2P)x2+1HS (6D+2P)x2+1HS (9D+2P)x2+1HS (12D+2P)x2+1HS (5D+2P)x4+1HS (13D+2P)x2+1HS (8D+2P)x3+1HS (4D+2P)x5+1HS (3D+2P)x6+1HS |
Reliability (RAID5+0) |
Multiple RAID5 volumes are RAID0 striped. For large capacity configurations, use of RAID5+0 instead of RAID5 results in enhanced performance, improved reliability, and shorter rebuilding times. Several restrictions apply to the RAID groups of this RAID level. Refer to "Restrictions for RAID5+0" for details. |
(2D+1P) × 2 - (15D+1P) × 2 |
Mirroring (RAID1) |
Data is mirrored to two drives (mirroring). If one drive fails, the other drive continues operation. |
1D+1M |
Striping (RAID0) |
Data is split in unit of blocks and stored across multiple drives (striping). RAID0 has no data redundancy. |
2D - 16D |
*1 | : | The RAID levels in the table above, setting value fields, and display contents fields for descriptions are written according to how they actually appear in Web GUI. Other fields and descriptions in this manual use "RAIDxx" as an abbreviation for the RAID levels. |
*2 | : | D: Data drives, M: Mirror drives, P: Parity drives, HS: Hot Spares |
Restrictions for RAID6-FR
The following restrictions apply to RAID groups (hereinafter referred to as "Fast Recovery RAID groups") that were created with RAID6-FR and volumes created in those RAID groups.
- The following operations that use LDE cannot be performed for Fast Recovery RAID groups.
Changing the RAID level to "RAID6-FR"
Changing the RAID level from "RAID6-FR"
Expanding the RAID group capacity by adding drives
"Standard (including concatenation volumes by means of LUN Concatenation)" and ODX Buffer volumes can be created in the Fast Recovery RAID group.
Encrypted volumes can be created in the Fast Recovery RAID groups. However, encrypting volumes that exist in Fast Recovery RAID groups is not allowed.
The Stripe Depth for a Fast Recovery RAID group is fixed at "64KB".
The Copybackless function is not performed when the first drive fails in a Fast Recovery RAID group. A high-speed rebuild is performed in a hot spare area within the RAID group, and when the failed drive is exchanged for a normal drive, copyback is performed.
Restrictions for RAID5+0
The following restrictions apply to RAID groups that are created with "RAID5+0".
- The following operations that use LDE cannot be performed.
Changing the RAID level to "RAID5+0"
Changing the RAID level from "RAID5+0"
Expanding the RAID group capacity by adding drives
The Stripe Depth for a new RAID group is fixed at "64KB".
The Maximum Number of RAID Groups for Each Model
The maximum number of RAID groups varies depending on each model. The following table shows the maximum number of RAID groups for each model.
Model | The maximum number of RAID groups |
---|---|
ETERNUS DX60 S5 |
48 |
ETERNUS DX100 S5 |
72 |
ETERNUS DX200 S5 |
132 |
ETERNUS DX500 S5 |
288 |
ETERNUS DX600 S5 |
528 |
ETERNUS DX900 S5 |
1152 |
ETERNUS DX8100 S4 | 24 |
ETERNUS DX8900 S4 | 3456 |
ETERNUS AF150 S3 |
12 |
ETERNUS AF250 S3 |
132 |
ETERNUS AF650 S3 |
528 |
Drive Combinations That Can Configure a RAID Group
The following table shows the drive combinations that can configure a RAID group.
Online | Nearline | SSD | Online SED | Nearline SED | SSD SED | |
---|---|---|---|---|---|---|
Online | OK | OK (but not recommended) | NG | NG | NG | NG |
Nearline | OK (but not recommended) | OK | NG | NG | NG | NG |
SSD | NG | NG | OK | NG | NG | NG |
Online SED | NG | NG | NG | OK | OK (but not recommended) | NG |
Nearline SED | NG | NG | NG | OK (but not recommended) | OK | NG |
SSD SED | NG | NG | NG | NG | NG | OK |
OK: RAID groups can be created OK (but not recommended): RAID groups can be created, but not a recommended configuration NG: RAID groups cannot be created
For the ETERNUS DX60 S5 and the ETERNUS AF150 S3, SEDs cannot be used for configuring a RAID group.
RAID0 has no data redundancy. The use of RAID1+0, RAID5, RAID6, RAID6-FR, RAID5+0, or RAID1 is recommended.
User Privileges
Availability of Executions in the Default Role
Default role | Availability of executions |
---|---|
Monitor | |
Admin | |
StorageAdmin | |
AccountAdmin | |
SecurityAdmin | |
Maintainer |
Refer to "User Roles and Policies" for details on the policies and roles.
Settings
In this screen, create a RAID group.
There are two methods to create RAID groups: automatic drive selection and manual drive selection.
New RAID Group
Input the name of the RAID group that is to be newly created and select the create mode.
Item | Description | Setting values |
---|---|---|
Name |
Input a RAID group name that is to be created. When creating a single RAID group, an existing RAID group name cannot be used. |
Up to 16 alphanumeric characters, symbols (except "," (comma) and "?"), and spaces |
Create Mode |
Select the create mode of the RAID group.
|
Automatic Manual |
Automatic Setting
Item | Description | Setting values |
---|---|---|
Number of RAID Groups |
Input the number of RAID groups that are to be created. When creating multiple RAID groups at a time, the new RAID groups are named automatically. Refer to "Naming Conventions for Creating RAID Groups" for details. |
For the ETERNUS DX60 S5: 1 - 48 For the ETERNUS DX100 S5: 1 - 72 For the ETERNUS DX200 S5: 1 - 132 For the ETERNUS DX500 S5: 1 - 288 For the ETERNUS DX600 S5: 1 - 528 For the ETERNUS DX900 S5: 1 - 1152 For the ETERNUS DX8100 S4: 1 - 24 For the ETERNUS DX8900 S4: 1 - 3456 For the ETERNUS AF150 S3: 1 - 12 For the ETERNUS AF250 S3: 1 - 132 For the ETERNUS AF650 S3: 1 - 528 0 (Default) |
Drive Type |
Select the type of drive that configures a RAID group from the list box. Only the drives that are installed in the storage system are displayed. Caution
|
|
RAID Level |
Select the level of RAID group that is to be created. Caution
|
High Performance (RAID1+0) High Capacity (RAID5) High Reliability (RAID6) High Reliability (RAID6-FR) Reliability (RAID5+0) Mirroring (RAID1) Striping (RAID0) |
Select Drives |
Select the requirements that are given priority when creating a Fast Recovery RAID group with automatic drive configuration. This item is available only when the RAID level is "RAID6-FR".
|
Minimize number of using drives Prioritize rebuild rate |
Minimum Capacity per RAID Group |
Input the RAID group capacity that is to be created and select the units of capacity. A RAID group is automatically created with a capacity of the entered value or higher. |
Numeric characters Unit: TB/GB/MB |
Naming Conventions for Creating RAID Groups
When creating multiple RAID groups at a time, a name is automatically added to a RAID group with the specified "Name" and a suffix number "x" (serial numbers starting with "0").
(Example) Specified RAID group name: RAIDGroup_aaaa (14 characters) → Name for created RAID group: RAIDGroup_aaaa0, RAIDGroup_aaaa1, etc.
When the RAID group name including the suffix number "x" has more than 16 characters, the excess number of characters is deleted from the "Name", starting with the last character and a suffix number "~x" will be added. Then, the name will contain only 16 characters.
(Example) Specified RAID group name: RAIDGroup_aaaabb (16 characters) → Name for created RAID group: RAIDGroup_aaaa~0, RAIDGroup_aaaa~1, etc.
When a RAID group name including the suffix number already exists, the suffix number is increased by one (+1). The suffix number is increased by one (+1) until no RAID group names overlap.
Manual Setting
Item | Description | Setting values |
---|---|---|
RAID Level |
Select the level of RAID group that is to be created. Caution
|
High Performance (RAID1+0) High Capacity (RAID5) High Reliability (RAID6) High Reliability (RAID6-FR) Reliability (RAID5+0) Mirroring (RAID1) Striping (RAID0) |
Controlling CM |
Specify the Controlling CM of the RAID group to be created. "Automatic" and the normal CM number ("CE#x CM#y" or "CM#y") that is installed are displayed as options. Select "Automatic" for normal operations. When "Automatic" is selected, the Controlling CM that is to be allocated is determined by the RAID group number. Refer to "Automatic Controlling CM Setting" for details. |
For the ETERNUS DX900 S5 or the ETERNUS DX8900 S4 Automatic CE#x CM#y For the other models Automatic CM#y x: CE number y: CM number |
Fast Recovery Configuration |
Select the drive configuration for a Fast Recovery RAID group. Select the drive configuration from "No. of drives in the configuration", "capacity efficiency", and "rebuilding speed" according to your environment. Refer to "Drive Configuration for Fast Recovery RAID Groups" for details. The more redundant sets there are, the faster the rebuilding becomes. This item is blank when the RAID level is not "RAID6-FR". |
For the ETERNUS DX60 S5 or the ETERNUS AF150 S3 (3D+2P)x2+1HS (4D+2P)x2+1HS (6D+2P)x2+1HS (9D+2P)x2+1HS Blank For the other models (3D+2P)x2+1HS (4D+2P)x2+1HS (6D+2P)x2+1HS (9D+2P)x2+1HS (12D+2P)x2+1HS (5D+2P)x4+1HS (13D+2P)x2+1HS (8D+2P)x3+1HS (4D+2P)x5+1HS (3D+2P)x6+1HS Blank D: Data drives P: Parity drives HS: Hot Spares |
Minimum Capacity per RAID Group |
The capacity of RAID group that is to be created is displayed. The "Minimum Capacity per RAID Group" is automatically calculated from the selected RAID level and drives. |
Drive Configuration for Fast Recovery RAID Groups
The drive layout in the storage system is the same as "RAID6". When using automatic configuration, the drive configuration that satisfies the specified capacity is determined according to the following order.
No. of drives in the configuration (per RAID group) |
Redundant sets + HS (*1) | Capacity efficiency (*2) (%) |
Rebuilding speed (*3) (Rate) |
Number of data drives | Selection order when configuring automatically | |
---|---|---|---|---|---|---|
"Minimize number of using drives" is selected | "Prioritize rebuild rate" is selected | |||||
11 | (3D+2P)x2+1HS | 54.5 | 2.20 | 6 | 1 | 5 |
13 | (4D+2P)x2+1HS | 61.5 | 2.17 | 8 | 2 | 6 |
17 | (6D+2P)x2+1HS | 70.6 | 2.13 | 12 | 3 | 7 |
23 | (9D+2P)x2+1HS | 78.3 | 2.09 | 18 | 4 | 8 |
29 | (12D+2P)x2+1HS | 82.8 | 2.07 | 24 | 5 | 9 |
31 | (13D+2P)x2+1HS | 83.9 | 2.06 | 26 | 6 | 10 |
31 | (3D+2P)x6+1HS | 58.1 | 6.20 | 18 | Not selected | 1 |
31 | (4D+2P)x5+1HS | 64.5 | 5.17 | 20 | Not selected | 2 |
29 | (5D+2P)x4+1HS | 70.0 | 4.14 | 20 | Not selected | 3 |
31 | (8D+2P)x3+1HS | 77.4 | 3.10 | 24 | Not selected | 4 |
*1 | : | Fast Recovery RAID groups are described as "Redundant sets + HS". RAID6 ((Number of data drives (D) + Number of parity drives (P)) × Number of redundant sets + Number of hot spares (HS)) ↑ Redundant sets (Example) "RAID6 ((3D+2P)x2+1HS)" is described as "(3D+2P)x2+1HS". |
*2 | : | The ratio of the user capacity to physical drive capacity. |
*3 | : | Rate when the rebuilding speed for the basic "RAID6 (D+P)" configuration is "1". The rate varies depending on the workload of the storage system and system environment. |
Advanced Setting
Perform the advanced settings for RAID groups
Item | Description | Setting values |
---|---|---|
Stripe Depth |
Stripe Depth should be selected only when advanced tuning needs to be performed for each RAID group. It is not necessary to change the default value for normal use. The setting is not available when the RAID level is "RAID1". Available Stripe Depth value varies depending on the RAID level. Refer to "Available Stripe Depth Value" for details. Note
|
64 KB (Default) 128 KB 256 KB 512 KB 1024 KB |
Available Stripe Depth Value
The Stripe Depth values available for each RAID level are as follows:
RAID level | Available Stripe Depth value |
---|---|
RAID1 | - |
RAID1+0, RAID0 | 64 KB, 128 KB, 256 KB, 512 KB, 1024 KB |
RAID5 (2+1) - RAID5 (4+1) | 64 KB, 128 KB, 256 KB, 512 KB |
RAID5 (5+1) - RAID5 (8+1) | 64 KB, 128 KB, 256 KB |
RAID5 (9+1) - RAID5 (15+1) | 64 KB, 128 KB |
RAID5+0 | 64 KB |
RAID6 | 64 KB |
RAID6-FR | 64 KB |
Restrictions for Stripe Depth Modification
Note that the following restrictions apply to a RAID group whose Stripe Depth has been changed and volumes created for such RAID group.
The Stripe Depth of the RAID groups already created cannot be changed .
When selecting drives automatically to create a RAID group, the Stripe Depth cannot be changed.
The capacity cannot be expanded if the Stripe Depth of the RAID group has been changed (LDE is not available).
Encryption cannot be performed for volumes that are created in a RAID group whose Stripe Depth has been changed.
Drive Selection
Drives can be selected from the list or the installation image. To switch between the list and the installation image, click the tab.
|
||||||
---|---|---|---|---|---|---|
|
[Tabular] Tab
Click the [Tabular] tab to select drives from the list. Only unused drives are displayed on the list.
There are conditions for the ETERNUS DX8100 S4/DX8900 S4 drive layout. Refer to "Conditions for the ETERNUS DX8100 S4/DX8900 S4 Drive Layout" for details. Note that these conditions are not applied to other models.
Item | Description | |||
---|---|---|---|---|
Checkbox to select drives |
Select the checkbox for the drive that is to be used. |
|||
Enclosure |
The enclosure where the drive is installed is displayed. CE: Controller Enclosure (2.5" and 3.5") DE: Drive Enclosure (2.5", 3.5", and 3.5" high density DEs) CE CE#x DE#yy x: CE number yy: DE number |
|||
Slot No. |
The slot number of the enclosure where the drive is installed is displayed. 2.5" CE/DE: 0 - 23 3.5" CE/DE: 0 - 11 3.5" high density DE: 0 - 59 |
|||
Type |
The drive type displayed for this item is a combination of the following.
Note that "SED" is also displayed for self encrypting drives and "AF" is also displayed for Advanced Format compliant drives.
|
|||
Capacity |
The capacity of the drive is displayed. Caution
|
|||
Speed |
The drive speed is displayed. For SSD or SSD SED, a "-" (hyphen) is displayed. 15000 rpm 10000 rpm 7200 rpm |
[Graphic] Tab
Click the [Graphic] tab to select drives from the drive installation image. The installation images of all the drives installed in the storage system are displayed. Checkboxes are displayed for unused drives.
There are conditions for the ETERNUS DX8100 S4/DX8900 S4 drive layout. Refer to "Conditions for the ETERNUS DX8100 S4/DX8900 S4 Drive Layout" for details. Note that these conditions are not applied to other models.
Item | Description | Setting values |
---|---|---|
DE selection list box |
Select the DE group. Options are displayed in the list box when at least one CE or DE in the DE group is installed in the storage system. Refer to "DE selection list box" for details on the options and DE groups for each model. |
DE#0x DE#1x DE#2x DE#3x DE#4x DE#5x DE#6x DE#7x DE#8x DE#9x DE#Ax DE#Bx DE#Cx DE#Dx DE#Ex DE#Fx |
DE |
Only the CEs or the DEs in the selected DE group that are installed in the storage system are displayed. CE CE#x DE#yy x: CE number yy: DE number |
|
Checkbox to select drives |
Select the checkbox for the drive that is to be used. Checkboxes are displayed for unused drives. For 2.5" CEs or 2.5" DEs, drives are displayed from left to right in ascending order of the slot number. For 3.5" CEs, 3.5" DEs, or 3.5" high density DEs, drives are displayed from bottom left to top right in ascending order of the slot number. Placing the mouse pointer on the icon displays the detailed information of the drive. |
Conditions for the ETERNUS DX8100 S4/DX8900 S4 Drive Layout
The drive layout to configure RAID groups in the ETERNUS DX8100 S4/DX8900 S4 must satisfy the conditions described below.
RAID groups cannot be created if the required conditions are not satisfied.
For the ETERNUS DX8100 S4
RAID level | Drive layout conditions | |
---|---|---|
RAID1 | Required | Allocate mirroring pair drives to different DEs. |
RAID1+0 | Required | Allocate mirroring pair drives to different DEs. |
Recommended | Allocate striping drives to as many DEs as possible. | |
RAID5 RAID5+0 RAID6 RAID6-FR |
Recommended | Distribute member drives to as many DEs as possible. |
For the ETERNUS DX8900 S4
RAID level | Drive layout conditions | |
---|---|---|
RAID1 | Required | Allocate mirroring pair drives to different DEs. |
Recommended | Allocate mirroring pair drives to DEs (*1) under different CEs when possible. Allocate mirroring pair drives to different SAS cascades (*2) when possible. |
|
RAID1+0 | Required | Allocate mirroring pair drives to different DEs. |
Recommended | Allocate striping drives to DEs under as many CEs as possible. Allocate striping drives to as many SAS cascades (*2) as possible. |
|
RAID5 | Required | Allocate member drives to different DEs. |
Recommended | Distribute member drives to DEs under as many CEs as possible. Distribute member drives to as many SAS cascades (*2) as possible. |
|
RAID5+0 | Required | Allocate two or less member drives to the same DE. Member drives in the same DE must belong to different redundant groups. |
Recommended | Distribute member drives to DEs under as many CEs as possible. Distribute member drives to as many SAS cascades (*2) as possible. |
|
RAID6 RAID6-FR |
Required | Allocate two or less member drives to the same DE. |
Recommended | Distribute member drives to DEs under as many CEs as possible. Distribute member drives to as many SAS cascades (*2) as possible. |
*1 | : | DEs under different CEs have different numbers as the first digit of the DE number. |
*2 | : | "SAS cascade" for the ETERNUS DX8900 S4 refers to DEs that are attached to one drive interface port. The DEs that are allocated to the same SAS cascade configuration are as follows: DE#x1, DE#x2, and DE#x3 that are connected to CE#x/DI Port#0 (x: 0 - B) DE#x4, DE#x5, DE#x6, and DE#x7 that are connected to CE#x/DI Port#1 (x: 0 - B) DE#x8, DE#x9, DE#xA, and DE#xB that are connected to CE#x/DI Port#2 (x: 0 - B) DE#xC, DE#xD, DE#xE, and DE#xF that are connected to CE#x/DI Port#3 (x: 0 - B) (Example) DE#01, DE#02, and DE#03 that are connected to CE#0/DI Port#0 are on the same SAS cascade. |
DE selection list box
Model | Option | DE group |
---|---|---|
ETERNUS DX60 S5 |
DE#0x | CE, DE#01 - DE#03 (for 3.5" DEs) |
ETERNUS DX100 S5 |
DE#0x | CE, DE#01 - DE#0A |
ETERNUS DX200 S5 |
DE#0x | CE, DE#01 - DE#0A |
ETERNUS DX500 S5 |
DE#0x | CE, DE#01 - DE#05 |
DE#1x | DE#10 - DE#15 | |
DE#2x | DE#20 - DE#25 | |
DE#3x | DE#30 - DE#35 | |
ETERNUS DX600 S5 |
DE#0x | CE, DE#01 - DE#0A |
DE#1x | DE#10 - DE#1A | |
DE#2x | DE#20 - DE#2A | |
DE#3x | DE#30 - DE#3A | |
ETERNUS DX900 S5 |
DE#0x | CE#0, DE#01 - DE#0F |
DE#1x | CE#1, DE#11 - DE#1F | |
DE#Cx | DE#C0 - DE#CF | |
DE#Dx | DE#D0 - DE#DF | |
DE#Ex | DE#E0 - DE#EF | |
DE#Fx | DE#F0 - DE#FF | |
ETERNUS DX8100 S4 | DE#0x | CE |
DE#1x | DE#10 | |
ETERNUS DX8900 S4 | DE#0x | CE#0 (*1), DE#01 - DE#0F |
DE#1x | CE#1 (*1), DE#11 - DE#1F | |
DE#2x | CE#2 (*1), DE#21 - DE#2F | |
DE#3x | CE#3 (*1), DE#31 - DE#3F | |
DE#4x | CE#4 (*1), DE#41 - DE#4F | |
DE#5x | CE#5 (*1), DE#51 - DE#5F | |
DE#6x | CE#6 (*1), DE#61 - DE#6F | |
DE#7x | CE#7 (*1), DE#71 - DE#7F | |
DE#8x | CE#8 (*1), DE#81 - DE#8F | |
DE#9x | CE#9 (*1), DE#91 - DE#9F | |
DE#Ax | CE#A (*1), DE#A1 - DE#AF | |
DE#Bx | CE#B (*1), DE#B1 - DE#BF | |
ETERNUS AF150 S3 |
DE#0x | CE |
ETERNUS AF250 S3 |
DE#0x | CE, DE#01 - DE#0A |
ETERNUS AF650 S3 |
DE#0x | CE, DE#01 - DE#0A |
DE#1x | DE#10 - DE#1A | |
DE#2x | DE#20 - DE#2A | |
DE#3x | DE#30 - DE#3A |
*1 | : | Only 2.5" drives can be installed. |
Operating Procedures
Automatically Selecting Drives to Create RAID Groups
Click [Create] in [Action].
Select "Automatic" for "Create Mode".
Specify the RAID group detailed information, and click the [Create] button.
→ A confirmation screen appears.
Caution- An error screen appears in the following conditions:
The "Name" overlaps with an existing name (when one RAID group is created)
The "Name" does not satisfy the input conditions
RAID groups cannot be created by using the drives that are installed in the storage system
- An error screen appears in the following conditions:
Click the [OK] button.
→ RAID group creation starts.
Click the [Done] button to return to the [RAID Group] screen.
Manually Selecting Drives to Create RAID Groups
Click [Create] in [Action].
Select "Manual" for "Create Mode".
Specify the RAID group detailed information.
Select drives using a list of the drives or the installation location image.
NoteWhen the number of drives for each RAID level and the number of selected drives does not match, the [Create] button cannot be clicked.
Click the [Create] button.
→ A confirmation screen appears.
Caution- An error screen appears in the following conditions:
The "Name" overlaps with an existing RAID group name
The "Name" does not satisfy the input conditions
The drive layout does not satisfy the required conditions
(Refer to "Conditions for the ETERNUS DX8100 S4/DX8900 S4 Drive Layout" for details.)
The specified Stripe Depth is not allowed for the RAID level
- An error screen appears in the following conditions:
Click the [OK] button.
→ RAID group creation starts.
Click the [Done] button to return to the [RAID Group] screen.