Storage controller features

The available features vary by storage controller model. For more information, see H3C Servers Storage Controllers Technical Specifications.

Hot spare drives

About hot spare drives

You can configure hot spare drives to improve data security. One or multiple independent drives in the drive system can function as hot spare drives. When a drive in a redundant RAID fails, a spare drive automatically replaces the failed drive and rebuilds the data of the failed drive, enhancing the fault tolerance of the system.

From the management interface or CLI of a storage controller, you can specify a hot spare drive for the RAID. Make sure the following conditions are met:

·     The hot spare drive is idle and of the same media type as a RAID member drive.

·     The capacity of the hot spare drive is equal to or larger than a RAID member drive.

Hot spare drives include the following types:

·     Global Spare—Specifies a hot spare drive for all the redundant RAIDs. You can specify one or multiple global spare drives. A global spare drive can replace any failed drive in any redundant RAID. When the storage controller detects that a new drive replaces the failed drive, states of the spare and new drives vary by support of the storage controller for data copyback:

¡     If the storage controller supports copying data on the global spare drive back to the new drive, the new drive becomes a RAID member drive and the global spare drive is placed in standby status.

¡     If the storage controller does not support copying data on the global spare drive back to the new drive, the global spare drive becomes a RAID member drive and the new drive is idle. You can configure the new drive as a hot spare drive from the management interface or CLI.

·     Dedicated Spare—Specifies a hot spare drive for a specific redundant RAID. You can specify one or multiple dedicated spare drives for a redundant RAID. A dedicated spare drive can replace a failed drive in the specific RAID. When the storage controller detects that a new drive replaces the failed drive, states of the spare and new drives vary by the support of the storage controller for data copyback:

¡     If the storage controller supports copying data on the dedicated spare drive back to the new drive, the new drive becomes a RAID member drive and the dedicated spare drive is placed in standby status.

¡     If the storage controller does not support copying data on the dedicated spare drive back to the new drive, the dedicated spare drive becomes a RAID member drive and the new drive is idle. You can configure the new drive as a hot spare drive from the management interface or CLI.

·     Auto Replace—Specifies a hot spare drive for a specific redundant RAID. Only P460 and H460 storage controllers support auto replace spare drives. You can specify one or multiple auto replace spare drives for a redundant RAID. An auto replace spare drive can replace a failed drive in the specific RAID. When the storage controller detects that a new drive replaces the failed drive, it will not copy data on the auto replace spare drive back to the new drive and place the new drive in standby status.

·     Pooled Spare—Specifies a hot spare drive for a RAID array set. You can specify one or multiple pooled spare drives for a RAID array set. A pooled spare drive can replace a failed drive in the RAID set. When the storage controller detects that a new drive replaces the failed drive, states of the spare and new drives vary by the support of the storage controller for data copyback:

¡     If the storage controller supports copying data on the pooled spare drive back to the new drive, the new drive becomes a RAID member drive and the pooled spare drive is placed in standby status.

¡     If the storage controller does not support copying data on the pooled spare drive back to the new drive, the pooled spare drive becomes a RAID member drive and the new drive is idle. You can configure the new drive as a hot spare drive from the management interface or CLI.

Applicable scenarios

The hot spare drive feature is applicable to critical data storage scenarios with high security and disaster recovery requirements, for example, unattended equipment rooms.

RAID rebuild

About RAID rebuild

RAID rebuild is required when a new drive or hot spare drive replaces a failed drive in a redundant RAID.

Applicable scenarios

The RAID rebuild feature is applicable to scenarios where automatic RAID rebuild is required after faulty drive replacement.

Copyback

About copyback

If a drive fails, the storage controller uses a hot spare drive to replace the failed drive and rebuilds data of the failed drive on the hot spare drive. When the storage controller detects that a new drive replaces the failed drive, it copies data on the hot spare drive back to the new drive and places the hot spare drive in standby status.

Applicable scenarios

If a drive fails, a hot spare drive will automatically replace the failed drive in a RAID. If the administrator uses a new drive to replace the failed drive, the administrator can set the new drive as a hot spare drive, changing the slot for the hot spare drive. With copyback enabled on a storage controller, the storage controller will copy data on the hot spare drive back to the new drive and places the hot spare drive in standby status.

Copyback ensures that hot spare drives remain in fixed slots for easy location and management.

Copyback affects drive performance. As a best practice, perform copyback when the system I/O is not busy.

RAID level migration

About RAID level migration

RAID level migration is to change the level of a RAID array to reconfigure RAID without compromising data integrity.

Applicable scenarios

You can perform RAID level migration when the current RAID level cannot meet service requirements.

RAID capacity expansion

About RAID capacity expansion

The storage controller supports the following methods to expand the capacity of a RAID array:

·     Adding available capacity—If the total capacity of all drives in a RAID array is not fully configured in RAID, the unconfigured capacity can be added to the RAID array to expand the RAID capacity.

·     Adding drives—New drives can be added to a RAID array to expand the RAID capacity. If needed, the level of the RAID array can be migrated to accommodate the newly added drives.

Applicable scenarios

You can perform RAID capacity expansion when the current RAID capacity cannot meet service requirements.

RAID location migration

About RAID location migration

This feature allows you to migrate logical drives in a RAID array or a whole RAID array to other physical drives with a sufficient capacity. The target physical drives can be of different types from the initial physical drives. For example, you can migrate logical drives on SATA HDDs to SAS SSDs.

Applicable scenarios

You can perform RAID location migration when bulk drive replacement, bulk data recovery, and host replacement occurs.

High speed read/write cache

About high speed read/write cache

The storage controller supports the cache feature. The cache provides a much faster read/write speed than drives.

·     Read policy—Monitoring data reads, the storage controller reads sequential data that will be requested in advance and stores the data in the cache. When the sequential data is requested, the controller directly reads the data stored in the cache. This mechanism speeds up reads for sequential data.

·     Write policy—The storage controller stores received data in the cache and notifies the system of the completion of data write. Without waiting for the data to be written into physical drives, the system immediately starts the next data transmission.

The storage controller writes data stored in the cache to physical drives when the data in the cache is accumulated to a certain amount.

Applicable scenarios

The high speed read/write cache feature is applicable to scenarios where high read/write speed is required.

RAID information clearing on drives

About RAID information clearing

If a drive is hot swapped, RAID information remains on the drive. You must clear RAID information on the drive before you configure the drive in a RAID array.

Applicable scenarios

The RAID information clearing feature is applicable to scenarios when you want to use a used drive to build a RAID array.

Power fail safeguard

About power fail safeguard

A power fail safeguard module is composed of a flash card and a supercapacitor. With write cache enabled, data stored in the cache gets lost if the server is powered down unexpectedly when power fail safeguard does not take effect. You can enable both write cache and power fail safeguard on the server to ensure data security:

·     Data in the cache will be copied to the flash card when the server is powered down unexpectedly.

·     Data in the flash card will be written to the drives when the server is powered on.

Applicable scenarios

You can configure power fail safeguard for critical data to provide data security.

Drive initialization and uninitialization

About drive initialization and uninitialization

Before RAID and hot spare drive configuration, some P430 storage controllers require drive initialization for drives in RAW status.

During initialization, the storage controller draws an area to save RAID configuration. During uninitialization, the storage controller clears all RAID configuration and places the drives in RAW status.

Applicable scenarios

Drive initialization is required if you use P430 storage controllers to configure RAID arrays. Drive uninitialization is required if you want to place the drives in RAW status.

Drive state switching

About drive state switching

You can change the state of drives attached to the storage controller.

PMC storage controllers with RAID capabilities support the following drive states:

·     RAW—Drives in this state cannot be used to configure RAID arrays but can be identified by the OS.

·     Ready—Drives in this state can be used to configure RAID arrays.

·     Optimal—Drives in this state have been used for configuring RAID arrays and are member drives of RAID arrays.

LSI storage controllers with RAID capabilities support the following drive states:

·     Unconfigured Good—Drives in this state are normal and can be used to configure RAID arrays or hot spare drives.

·     Unconfigured Bad—Drives in this state are faulty or have RAID information remaining on them. If a drive is faulty, replace the drive. If RAID information remains on a drive, clear the RAID information.

·     Hotspare—Drives in this state act as hot spare drives.

·     JBOD—Just a Bunch Of Drives. Drives in this state cannot be used to configure RAID arrays.

Applicable scenarios

You can change the state of drives as required.

Drive locating

About drive locating

This feature allows the storage controller to turn the Fault/UID LEDs on the drives to steady blue, which makes it easier for you to find the locations of the drives.

Applicable scenarios

To manually locate the drives, you can use this feature.

Logical drive initialization

About logical drive initialization

You must initialize logical drives after you configure a RAID array. After logical drives are initialized, they can be used by the operating system and the member drives in the redundant RAID array can meet the requirement of the RAID level.

The storage controller supports the following methods to initialize a logical drive:

·     Fast Initialize Drive—Firstly initializes the first 10 MB and the last 10 MB of the logical drive, which sets the RAID state to Optimal.

·     Slow Initialize Drive—Initializes all space in the logical drive. Data write is available after the initialization is finished.

Applicable scenarios

To make the member drives in the redundant RAID array meet the requirement of the RAID level, you can use this feature.

Drive scanning

About drive scanning

You can perform this task for the storage controller to detect the status change of a hot-swapped drive timely.

Applicable scenarios

You can use this feature to manually refresh the drive information.

Drive erasing

About drive erasing

CAUTION: ·     If the operating system is installed on a logical drive, erasing the logical drive might cause system failure. Perform this operation with caution. ·     To avoid damage to drives, do not perform any other operations while the controller is erasing drives.

 

This feature enables the storage controller to completely delete data on drives and prevents data from being recovered. You can perform this task to protect data security.

The drive erasing operation might take a few hours and cannot be aborted deliberately. If this operation is aborted due to server powering off or reboot, drives might get damaged or become unusable.

 

	NOTE: For some storage controller, this feature is available only when the storage controller operates in RAID mode.

 

Applicable scenarios

To completely delete data on drives and prevent data from being recovered, you can use this feature on storage controllers. Before you use this feature, make sure the storage controllers support this feature.

Drive failure alarming

About drive failure alarming

When a storage controller detects a failing drive, HDM generates an alarm and turns the Fault/UID LED on the drive backplane to flashing amber.

When a storage controller detects a failed drive, the storage controller turns the Fault/UID LED on the drive backplane to steady amber.

Applicable scenarios

You can use this feature to locate failed or failing drives.

Online firmware upgrade

About online firmware upgrade

The storage controller supports updating the firmware online from the CLI.

Applicable scenarios

You can use this feature to update the firmware of storage controllers.

Drive hot-swapping

About drive hot-swapping

CAUTION: To avoid drive damage, make sure there is a minimum of 30 seconds between two consecutive drive hot swapping operations.

 

The storage controller allows you to hot swap drives. You can remove or replace failed drives without shutting down the server system or powering off the server. This feature improves disaster recovery capability and capacity scalability of the server.

Applicable scenarios

You can use this feature to replace drives without powering off the server.

RAID creation with NVMe drives

About RAID creation with NVMe drives

You can use Intel VROC or specific storage controllers to create RAIDs with NVMe drives to meet performance, cost efficiency, and high availability requirements.

Applicable scenarios

This feature is applicable to scenarios when you need to create RAIDs with NVMe drives.