The approximate processing time [min] until drive data relocation is completed can be calculated as follows:
For HPEC 4D + 1P:
Processing time [min] = ((Total physical capacity of drives of expanded storage pool installed in the added or removed storage nodes - 625 [GiB]) / 625 [GiB]) × unit processing time for drive data relocation [min]
For HPEC 4D + 2P:
Processing time [min] = ((Total physical capacity of drives of expanded storage pool installed in the added or removed storage nodes - 900 [GiB]) / 900 [GiB]) × unit processing time for drive data relocation [min]
For Mirroring Duplication:
Processing time [min] = ((A – 200 [GiB]) / 200 [GiB]) × B [min]
(A: Total physical capacity of drives that are added to an expanded storage pool and are installed on the storage node to be added or removed, B: Unit processing time for drive data relocation)
If physical capacity of a storage node exceeds the capacity that can be allocated to a storage controller, the calculated approximate processing time might be longer. For details about the maximum capacity that can be allocated to a storage controller, contact customer support
Depending on the user data protection method (redundantType) settings and resource usage rate of internal processing I/O (asyncProcessingResourceUsageRate), unit processing time [min] for drive data relocation is as follows.
The asyncProcessingResourceUsageRate setting determines the usage of internal processing I/O (rebuild and drive data relocation) resources. To change the setting, see Changing usage of internal processing I/O resources.
|
redundantType |
asyncProcessingResourceUsageRate |
Unit processing time for drive data relocation (approximate time) [min] |
|---|---|---|
|
4D+1P |
VeryHigh/High/Middle |
100 |
|
Low |
200 |
|
|
4D+2P |
VeryHigh/High/Middle |
300 |
|
Low |
500 |
|
|
Duplication |
VeryHigh/High/Middle |
11 |
|
Low |
24 |
-
Setting asyncProcessingResourceUsageRate to VeryHigh, High, or Middle makes no difference in the processing time of drive data relocation.
-
Processing time [min] for drive data relocation is based on measurement results in a configuration that meets the following prerequisites:
-
With I/O load
-
Number of installed user data drives per storage node: 6
-
User data drive to be used: SAS SSD
-
Internode network bandwidth: 10 Gbps
-
MTU size of the network switch: Set to 9000
-
The rebuild capacity policy (rebuildCapacityPolicy) is set to "Fixed" and the tolerable number of drive failures (numberOfTolerableDriveFailures) is set to 1 (default). Alternatively, the rebuild capacity policy is set to "Variable."
-
The processing time [min] for drive data relocation might increase or decrease from the approximate time due to the following factors.
-
I/O load conditions.
-
If the number of installed user data drives per storage node is less than six, the processing time might be longer than the approximate time.
(Due to increased I/O load on each user data drive)
-
When the user data protection method is HPEC and storage nodes are removed.
No less than the approximate processing time and no more than approximately the following time are required.
-
4D+1P: approximate processing time [min] × (number of storage nodes before removal / 5)
-
4D+2P: approximate processing time [min] × (number of storage nodes before removal / 6)
-
-
When the rebuild capacity policy is set to "Fixed"
-
If the tolerable number of drive failures is 0, the processing time might be longer than the approximate time.
-
If the tolerable number of drive failures is two or more, the processing time might be shorter than the approximate time.
-