Selecting a Capacity Expansion Method
An appropriate capacity expansion method depends on factors such as the amount of required capacity expansion, performance improvement, and reliability improvement.
Capacity Expansion Methods
Storage systems can be expanded by adding:
- Disks
- Disk enclosures
- Disk bays (applicable to 18000 V5 and 18000F V5)
- Controllers
- Links
You can select an appropriate capacity expansion method based on storage system conditions and capacity requirements of services running on the storage systems.
Table 3-1 describes the characteristics and application scenarios of each capacity expansion method.
Capacity Expansion Method |
Characteristics |
Application Scenario |
|---|---|---|
Adding disks |
|
The storage system has enough empty disk slots to satisfy your capacity expansion requirements. |
Adding disk enclosures |
|
The storage system does not have enough empty disk slots or the total capacity of the empty disk slots is insufficient for your capacity expansion requirements. For example, 5 TB of capacity is required but only 2 TB is available after disks are added to all free disk slots. |
Adding disk bays |
|
The total capacity of fully loaded existing bays is insufficient for your capacity expansion requirements. |
Adding controllers |
|
The storage system does not meet service performance requirements. |
Adding links |
|
The storage system does not meet service reliability requirements. |
For a storage system with two or more controller enclosures, the disk enclosures that have been identified by a controller enclosure cannot be connected to any other controller enclosure of the same storage system after removal. Otherwise, the storage system cannot be powered on.
Capacity Expansion Roadmap
- Figure 3-1 shows the capacity expansion roadmap for 5000 V5, 5000F V5, 6000 V5, and 6000F V5.
- Figure 3-2 shows the capacity expansion roadmap for 18000 V5 and 18000F V5.
- Planned expansion capacity refers to the capacity that you want to add to the system. This capacity is determined from the collected live network information and the workload of ongoing services. Planned expansion capacity must include capacity used by the file system, RAID policy, hot spare policy, and metadata. This capacity must be sufficient for storage requirements of future data growth. For details about planned expansion capacity, consult your service provider.
The planned expansion capacity recommended by your service provider is usually the one identified by an application server.
- The allowed expansion capacity is the maximum capacity that can be expanded using existing free disk slots.Allowed expansion capacity = Capacity of a disk x Number of free disk slots
- You can estimate the actual capacity of a disk based on the capacity of existing disks used in the storage system.
- Use the actual capacity instead of the nominal capacity.
Disk manufacturers use 1000 as the conversion factor: 1 GB = 1000 MB, 1 MB = 1000 KB, 1 KB = 1000 bytes
Operating systems use 1024 as the conversion factor: 1 GB = 1024 MB, 1 MB = 1024 KB, 1 KB = 1024 bytes
The storage systems provide the data integrity field (DIF) function for end-to-end data protection. This function takes 1% to 2% of storage space.
For example, when 600 GB SAS disks are used for the current network, the expansion capacity is 559 GB (rather than 600 GB) x (98% to 99%) x Number of available disk slots. The disk capacity defined by disk manufacturers is different from that calculated by operating systems. Therefore, the nominal capacity of a disk is different from that displayed in the operating system.
The disk capacity displayed in the operating system is for reference only. The disk capacity reported in the DeviceManager management software always prevails.
- For details about how to calculate the allowed expansion capacity, see Calculating the Allowed Expansion Capacity.
Calculating the Allowed Expansion Capacity
The allowed expansion capacity is equal to the planned expansion capacity minus the capacity used by the RAID policy, hot spare policy, and metadata.
If you want to expand the system capacity by adding controllers, consider the coffer disk capacity when calculating the allowed expansion capacity.
- Capacity used by parity data or mirrored data in a RAID group
Table 3-2 lists the disk utilization of different RAID levels.
Table 3-2 Disk utilization of different RAID levelsRAID Level
Disk Utilization
RAID 0
100%
RAID 1
- 2Da: about 50%
- 4D: about 25%
RAID 3
RAID 3 supports flexible configurations. Specifically, a RAID 3 policy allows data block and parity block policies ranging from 2D+1P to 13D+1P. The following examples show disk utilization of several configurations commonly used by RAID 3:
- 4D+1Pb: about 80%
- 2D+1P: about 66.67%
- 8D+1P: about 88.89%
RAID 5
RAID 5 supports flexible configurations. Specifically, a RAID 5 policy allows data block and parity block policies ranging from 2D+1P to 13D+1P. The following examples show disk utilization of several configurations commonly used by RAID 5:
- 2D+1P: about 66.67%
- 4D+1P: about 80%
- 8D+1P: about 88.89%
RAID 6
RAID 6 supports flexible configurations. Specifically, a RAID 6 policy allows data block and parity block policies ranging from 2D+2P to 26D+2P. The following examples show disk utilization of several configurations commonly used by RAID 6:
- 2D+2P: about 50%
- 4D+2P: about 66.67%
- 8D+2P: about 80%
- 16D+2P: about 88.89%
RAID 10
50%
RAID 50
- (2D+1P) x 2: about 66.67%
- (4D+1P) x 2: about 80%
- (8D+1P) x 2: about 88.89%
a: D indicates a data block.
b: P indicates a parity block.
NOTE:For a flexibly configured RAID policy xD+yP, the disk utilization is [x/(x + y)] x 100%.
- Capacity used by hot spare space
To prevent data loss or performance deterioration caused by a member disk failure, a storage system uses hot spare space to take over data from the failed member disk. The following hot spare policies are supported:
- 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. The hot spare space non-linearly increases as the number of disks increases.
- 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. The hot spare space non-linearly increases as the number of disks increases.
Number of disks of which capacity is used as hot spare space in a low hot spare policy = Number of disks of which capacity is used as hot spare space in a high hot spare policy/2 (rounded up)
- None
The storage system does not provide hot spare space.
Table 3-3 describes how hot spare space changes for a single engine with the number of disks. The hot spare space changes at a storage tier are used as an example here. The hot spare space changes at different storage tiers are the same.
Table 3-3 Changes of hot spare space for a single engineNumber of Disks
Number of Disks of Which Capacity Is Used as Hot Spare Space in a High Hot Spare Policya
Number of Disks of Which Capacity Is Used as Hot Spare Space in a Low Hot Spare Policya
(1, 12]
1
1
(12, 25]
2
(25, 125]b
3
2
(125, 325]
4
……
a: Huawei storage systems use RAID 2.0+ virtualization technology. Hot spare capacity is provided by member disks in each disk domain. Therefore, the hot spare capacity is expressed in the number of disks in this table.
For example, if a disk domain is composed of 12 SSDs and the high hot spare policy is used, the hot spare space occupies the capacity of one SSD and the capacity is provided by member disks in the disk domain. If a disk domain is composed of 13 SSDs and the high hot spare policy is used, the hot spare space occupies the capacity of two SSDs.
b: When the number of disks at a storage tier reaches 125, the storage tier uses the capacity of one disk in every 200 additional disks as the hot spare space in a high hot spare policy.
- Number of Disks in the above tables refers to the number of same-type disks owned by a same engine. If you select disks from multiple engines to create a disk domain, calculate the number of disks used for hot spare space on each engine and sum up the values.
- 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 a specific disk domain only.
- Common capacity changes of the hot spare space are listed in this section. The number of disks supported by a storage system and the capacity of hot spare space are based on actual specifications.
- High
- Capacity used by coffer disks
Part of the coffer disk space is used to store critical system data, including configuration data and system logs. The rest of the coffer disk space is used to store service data.
For details about capacity used by coffer disks, see "Coffer Disk" in the Product Description specific to your product model and version.
- Capacity used by file systems and volume management software on the application server
File systems and volume management software on the application server may occupy capacities in the storage system. The actually occupied capacities depend on the deployment of applications on the application server.
- WriteHole capacity
WriteHole is used to resolve inconsistent data stripe verification caused by certain operations before I/Os are delivered to disks. Each disk reserves a 256 MB space as WriteHole capacity.
- Capacity used by system information
The system information occupies 577 MB per disk.
- Metadata capacity
Each disk reserves 0.6% of its total capacity as metadata capacity, and reserves 2% as metadata backup capacity.
- Capacity reserved for improving system performance and disk balance
Each disk reserves 1% of its total capacity to improve system performance and disk balance. When 1% of the disk total capacity is smaller than 2 GB, 2 GB capacity is reserved.
- Integrated capacity
When disks are being formatted, if the size of a sector is 520 bytes, the sector uses 8 bytes to store parity data. If the size of a sector is 4160 bytes, the sector uses 64 bytes to store parity data. The integrated capacity usage is about 98.46% (512/520 or 4096/4160).
Without considering the hot spare capacity consumption, you can use the following formula to calculate RAID 2.0+ disk capacity usage: RAID 2.0+ disk capacity usage = [1 – Metadata space – (1 – Metadata space) × Metadata backup space] × (1 – Disk space reserved for load balancing) x Integrated capacity usage = [1 – 0.6% – (1 – 0.6%) × 2%] × (1 – 1%) × 98.46% ≈ 94.95%
The disk capacity defined by disk manufacturers is different from that calculated by operating systems. As a result, the nominal capacity of a disk is different from that displayed in the operating system.
- Disk capacity defined by disk manufacturers: 1 GB = 1,000 MB, 1 MB = 1,000 KB, 1 KB = 1,000 bytes.
- Disk capacity calculated by operating systems: 1 GB = 1,024 MB, 1 MB = 1,024 KB, 1 KB = 1,024 bytes.
Available Capacity Calculation Method
The following uses an example to explain how to calculate the allowed available capacity. Three valid digits are retained after the decimal point.
Assume that forty-eight 600 GB SAS disks will be added to the storage system, including four coffer disks and the hot spare policy and RAID policy are configured to Low and RAID 6 (8D + 2P) respectively. The allowed available capacity is calculated as follows:
- 600 GB is the nominal capacity provided by the disk manufacturer. Use the following method to convert this capacity to one that can be identified by the storage system:
600 GB x 1000 x 1000 x 1000/1024/1024 = 572204.590 MB
- Minus the WriteHole capacity:
572204.590 MB – 256 MB = 571948.590 MB
- Minus the capacity used by system information:
571948.590 MB – 577 MB = 571371.590 MB
- Minus the metadata capacity:
571371.590 MB x (1 – 0.6%) = 567943.361 MB
The storage system reserves 0.6% of each disk's space as metadata space. It dynamically allocates metadata space as services increase. The actual services prevail. In this example, 0.6% is used.
- Minus the metadata backup capacity:
567943.361 MB x (1 – 2%) = 556584.494 MB
- Minus the capacity reserved for improving system performance and disk balance:
556584.494 MB x (1 – 1%) = 551018.649 MB
- Minus the integrated capacity:
551018.649 MB x 98.46% = 542532.962 MB
- Minus the hot spare space capacity. Because the hot spare policy of the storage system is set to Low, capacity of two disks is used as hot spare space capacity. Therefore, the remaining capacity is as follows after the hot spare space capacity is deducted:
542532.962 MB x (48 – 2) = 24956516.250 MB
Equals to 24956516.250 MB/1024/1024 = 23.800 TB
- Minus the capacity used by coffer disks:
23.800 TB – 4 x 5 GB = 23.781 TB
In this example, the capacity used by each coffer disk is calculated as 5 GB. For details about capacity used by coffer disks, see "Coffer Disk" in the Product Description specific to your product model and version.
- Because the RAID policy of the storage system is RAID 6 (8D+2P), the disk utilization is 80%. Therefore, the allowed available capacity is:
23.781 TB x 80% = 19.025 TB
In this example, the allowed available capacity is 19.025 TB.
The preceding available capacity is for reference only. The capacity displayed on DeviceManager prevails.
The evaluation results of the local Huawei representative office or Huawei authorized distributor supersede the considerations in this section.
