No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

What Is Hot Spare Space

Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
What Is Hot Spare Space

What Is Hot Spare Space

Concepts

A disk array uses hot spare space to improve system reliability.

What Is Hot Spare in a Disk Array?

When a disk in a disk array is faulty, the disk array restores data in the faulty disk to the reserved space. This mechanism is hot spare and the reserved space is hot spare space. Hot spare space does not store any user data. Hot spare space can come from a physical disk or the space of each member disk in the same layer of a disk domain by using the virtualization technology.

Purpose of Hot Spare

To prevent performance deterioration caused by a member disk failure, a disk array employs hot spare space to take over data from the failed member disk.

What Is Traditional Hot Spare?

Traditionally, several idle disks are reserved as hot spare disks to provide hot spare space.

Huawei OceanStor T series V1 storage arrays use this traditional way. These reserved disks are global hot spare disks. They cannot be used as the hot spare disks for specific RAID groups. When a disk in a RAID group is faulty, data on the faulty disk is reconstructed on a hot spare disk. After the faulty disk is replaced, data on the hot spare disk is automatically restored to the new disk.

What Is Hot Spare in RAID 5?

Some people think that the RAID level is related to the size of hot spare space. In fact, they are not directly related.

RAID 5 is a RAID level that balances storage performance, data security, and costs. RAID 5 implements parity check and distributes parity data on all member disks. When a member disk is faulty, its data can be recovered based on the data on the other member disks.

As shown in the following figure, P0 is the parity data of stripes D0, D1, and D2, P1 is the parity data of stripes D3, D4, and D5, and the rest can be deduced by analogy. RAID 5 does not directly back up the data stored. Instead, data and its parity information are stored on different member disks in the array. If data on a member disk is damaged, the data can be restored from the remaining data and its parity information. If data on a RAID 5 member disk is damaged, RAID 5 can use the remaining data and the corresponding parity information to restore the damaged data.

Figure 1-1 Data distribution on a RAID 5 disk array

Assuming that there are N disks in a RAID group, when RAID 5 is used, the number of disks storing user data in the RAID group is N-1 and the number of disks storing parity data is 1. Hot spare space functions as an additional disk to the member disks of the RAID group. The hot spare space does not affect the available capacity of the RAID group. With hot spare, the number of disks used to store user data in the RAID group is still N-1. Therefore, no matter whether there is hot spare, a maximum of one disk can be damaged when RAID 5 is used.

What Is Hot Spare in RAID 10?

Some people think that the RAID level is related to the size of hot spare space. In fact, they are not directly related.

RAID 10 is a combination of RAID 0 and RAID 1. It allows drives to be mirrored (RAID 1) and then striped (RAID 0). RAID 10 is a solution that provides good storage performance (similar to RAID 0) and data security (same as RAID 1).

As shown in Figure 1-2, disks 0 and 1 form subgroup 0, disks 2 and 3 form subgroup 1, and disks in the same subgroup are mirrors of each other. If I/O requests are sent to disks in RAID 10, the sequential data requests are distributed to the two subgroups for processing (RAID 0 mode). At the same time, in RAID 1 mode, when data is written into disk 0, a copy is created on disk 1; when data is written into disk 2, a copy is created on disk 3.

Figure 1-2 Data distribution on a RAID 10 disk array

Assuming that there are N disks in a RAID group, when RAID 10 is used, the number of disks storing user data in the RAID group is N/2 and the number of disks storing parity data is N/2. Hot spare space functions as an additional disk to the member disks of the RAID group. The hot spare space does not affect the available capacity of the RAID group. With hot spare, the number of disks used to store user data in the RAID group is still N/2. Therefore, no matter whether there is hot spare, a maximum of N/2 disks can be damaged when RAID 10 is used.

What Is Hot Spare in Huawei RAID2.0+?

Huawei RAID2.0+ technology is a new RAID technology designed for new-generation storage arrays (virtualization, hybrid cloud, simplified IT, and low-carbon) to meet the development trend of storage technology virtualization architectures.

RAID2.0+ uses the block virtualization technology to dynamically allocate resources in a storage pool, improving data read/write speed and shortening the reconstruction time after a disk fault occurs.

Figure 1-3 Block virtualization process
  1. All disks in a storage array are divided into disk domains. Each disk domain can consist of multiple disks of the same or different types. The disk types of a disk domain determine the storage tiers of a storage pool.
  2. The storage array divides the storage medium at each storage tier into chunks of a fixed size.
  3. Chunks at each storage tier form chunk groups and hot spare space based on the RAID policy and hot spare policy specified in DeviceManager.

RAID2.0+ automatically reserves certain hot spare space in a disk domain based on the hot spare policy. When a chunk on a disk is faulty, the system uses the hot spare space to replace the faulty chunk, in addition, it notifies other chunks in the chunk group to reconstruct data in the hot spare space, thereby implementing rapid self-healing and ensuring data security and read/write performance.

RAID 2.0+ and RAID 1.0 have a big difference in hot spare. RAID 1.0 uses dedicated hot spare disks which are idle when they are not participating in reconstruction. RAID 2.0+ does not use dedicated hot spare disks. The hot spare space is distributed on all member disks and is an independent "idle" space. Reconstruction in RAID 2.0+ is more than ten times faster than that in RAID 1.0. In RAID 1.0, only one disk participates in reconstruction. In RAID 2.0+, all disks in the disk domain participate in reconstruction.

How Do I Calculate Hot Spare Space (Huawei RAID 2.0+)?

How Do I Calculate the Hot Spare Space of an OceanStor V3/V5 Series Storage Array?

Huawei OceanStor V3/V5 series storage arrays support the following hot spare policies:
  • High

    The capacity of one disk is used as hot spare space if the number of disks at a storage tier is equal to or fewer than 12. With the increase of the number of disks, the hot spare space increases non-linearly.

  • Low

    The capacity of one disk is used as hot spare space if the number of disks at a storage tier is equal to or fewer than 25. With the increase of the number of disks, the hot spare space increases non-linearly.

    Number of disks for hot spare in the low hot spare policy = Number of disks for hot spare in the high hot spare policy/2 (rounded up)

  • None (OceanStor 18000/18000F series storage arrays do not provide hot spare.)

    When a member disk in a disk domain fails, the system uses the free capacity of the disk domain for reconstruction. If the free capacity of the disk domain is insufficient, the system uses the unallocated space in the storage pool for reconstruction. If the reconstruction fails, the disk domain will be degraded, the read/write performance will deteriorate, and the reliability of the storage system will be affected.

    For V300R006C50, V500R007C30, and earlier versions, the size of hot spare space (single engine) varies with the number of disks, as described in Table 1-1. A same storage tier is used as an example here. This relationship applies to different types of storage tiers.

    Table 1-1 Relationship between the number of disks and the size of hot spare space (single engine)

    Number of Disks

    Number of Disks for Hot Spare in High Hot Spare Policya

    Number of Disks for Hot Spare in Low Hot Spare Policya

    (1, 12]

    1

    1

    (12, 25]

    2

    (25, 50]

    3

    2

    (50, 75]

    4

    (75, 125]

    5

    3

    (125, 175]b

    6

    (175, 275]

    7

    4

    (275, 375]

    8

    a: Hot spare space is distributed on member disks at the same tier of a disk domain based on RAID 2.0+ and underlying virtualization technologies. For ease of understanding, the size of hot spare space is expressed in the number of disks here.

    For example, if a disk domain is composed of 12 SSDs and the high hot spare policy is used, hot spare space is provided by member disks in the disk domain but the size of the hot spare space equals to the capacity of one SSD. If a disk domain is composed of 13 SSDs and the high hot spare policy is used, the size of hot spare space equals to the capacity of two SSDs. Other situations can be deduced by analogy.

    b: When the number of disks is more than 175 and the high hot spare policy is used, the capacity of one disk is added as hot spare space for every 100 new disks at the storage tier.

    For V300R006C60, V500R007C50, and later versions, the size of hot spare space (single engine) varies with the number of disks, as described in Table 1-2. A same storage tier is used as an example here. This relationship applies to different types of storage tiers.

    Table 1-2 Relationship between the number of disks and the size of hot spare space (single engine)

    Number of Disks

    Number of Disks for Hot Spare in High Hot Spare Policya

    Number of Disks for Hot Spare in Low Hot Spare Policya

    (1, 12]

    1

    1

    (12, 25]

    2

    (25, 125]b

    3

    2

    (125, 325]

    4

    a: Hot spare space is distributed on member disks at the same tier of a disk domain based on RAID 2.0+ and underlying virtualization technologies. For ease of understanding, the size of hot spare space is expressed in the number of disks here.

    For example, if a disk domain is composed of 12 SSDs and the high hot spare policy is used, hot spare space is provided by member disks in the disk domain but the size of the hot spare space equals to the capacity of one SSD. If a disk domain is composed of 13 SSDs and the high hot spare policy is used, the size of hot spare space equals to the capacity of two SSDs. Other situations can be deduced by analogy.

    b: When the number of disks is more than 125 and the high hot spare policy is used, the capacity of one disk is added as hot spare space for every 200 new disks at the storage tier.

    NOTE:
    • For a disk domain created in a version earlier than V300R006C60 or V500R007C50 but later the storage array is upgraded to V300R006C60, V500R007C50, or later, the hot spare space of the disk domain is recalculated according to Table 1-2.
    • The number of disks listed on the left of the table refers to the number of disks of the same type from the same engine. If the disks of a disk domain come from multiple engines, the number of disks for hot spare space needs to be calculated for each engine and then summed.
    • For OceanStor 18000 series/18000F series storage arrays, the high hot spare policy is used by default. Hot spare policies cannot be configured for specific storage tiers on DeviceManager. You can run the change disk_domain general command on the CLI to modify the default policy.
    • When you are creating a disk domain, ensure that the disks used to provide hot spare space are sufficient.
    • Hot spare space can be used for the current disk domain only.
    • Only common sizes of hot spare space are listed here. The number of disks supported by a storage array and the size of hot spare space are subject to the actual situation.

How Do I Calculate the Hot Spare Space of an OceanStor Dorado V3 Series Storage Array?

OceanStor Dorado V3 series storage arrays support hot spare to take over data on failed member disks to ensure the system continuity and reliability. The supported hot spare policies are as follows:

  • High

    The capacity of one disk is used for hot spare when the system has up to 12 disks. With the increase of the number of disks, the hot spare space increases non-linearly.

  • Low

    The capacity of one disk is used for hot spare when the system has up to 25 disks. With the increase of the number of disks, the hot spare space increases non-linearly.

  • None

    The arrays do not provide hot spare space. When a member disk in a disk domain fails, the system uses the free capacity of the disk domain for reconstruction. If the free capacity of a disk domain is insufficient, the system will occupy the space with no data written into in the storage pool for reconstruction. If the reconstruction fails, the disk domain will be degraded, the read/write performance will deteriorate, and the reliability of the storage system will be affected.

    In V300R002C00, the size of hot spare space for each pair of controllers varies with the number of disks, as shown in Table 1-3.

    Table 1-3 Relationship between the number of disks and the size of hot spare space

    Number of Disks

    Number of Disks for Hot Spare in High Hot Spare Policya

    Number of Disks for Hot Spare in Low Hot Spare Policya

    [8,12]

    1

    1

    (12,25]

    2

    (25,50]

    3

    2

    (50,75]

    4

    (75,125]

    5

    3

    (125,175]

    6

    (175,275]

    7

    4

    (275,300]

    8

    4

    a: Hot spare space is distributed on member disks at the same tier of a disk domain based on RAID 2.0+ and underlying virtualization technologies. For ease of understanding, the size of hot spare space is expressed in the number of disks here.

    For example, if a disk domain is composed of 12 SSDs and the high hot spare policy is used, hot spare space is provided by member disks in the disk domain but the size of the hot spare space equals to the capacity of one SSD. If a disk domain is composed of 13 SSDs and the high hot spare policy is used, the size of hot spare space equals to the capacity of two SSDs. Other situations can be deduced by analogy.

    NOTE:
    • When you are creating a disk domain, ensure that the disks used to provide hot spare space are sufficient.
    • Hot spare space can be used for the current disk domain only.
    • Table 1-3 lists common sizes of hot spare space. The number of disks supported by a storage array and the size of hot spare space are subject to the actual situation.

    In V300R002C10 and later versions, the size of hot spare space for each pair of controllers varies with the number of disks, as shown in Table 1-4.

    Table 1-4 Relationship between the number of disks and the size of hot spare space

    Number of Disks

    Number of Disks for Hot Spare in High Hot Spare Policya

    Number of Disks for Hot Spare in Low Hot Spare Policya

    [8,12]

    1

    1

    (12,25]

    2

    (25,75]

    3

    2

    (75,300]

    3

    a: Hot spare space is distributed on member disks at the same tier of a disk domain based on RAID 2.0+ and underlying virtualization technologies. For ease of understanding, the size of hot spare space is expressed in the number of disks here.

    For example, if a disk domain is composed of 12 SSDs and the high hot spare policy is used, hot spare space is provided by member disks in the disk domain but the size of the hot spare space equals to the capacity of one SSD. If a disk domain is composed of 13 SSDs and the high hot spare policy is used, the size of hot spare space equals to the capacity of two SSDs. Other situations can be deduced by analogy.

    NOTE:
    • When you are creating a disk domain, ensure that the disks used to provide hot spare space are sufficient.
    • Hot spare space can be used for the current disk domain only.
    • Table 1-4 lists common sizes of hot spare space. The number of disks supported by a storage array and the size of hot spare space are subject to the actual situation.
    • For a storage array that is upgraded from V300R002C00 or an earlier version to V300R002C10 or later, the hot spare space of the disk domain created before the upgrade is calculated based on the rules of the source version and the hot spare space of the disk domain created after the upgrade is calculated based on the rules of target version.
Download
Updated: 2019-08-07

Document ID: EDOC1100092550

Views: 456

Downloads: 32

Average rating:
This Document Applies to these Products
Related Documents
Related Version
Share
Previous Next