PERC 5 Controller Features
Dell™ PowerEdge™ Expandable RAID Controller 5/i and 5/E User's Guide
RAID Configuration Information
This section describes the features of the The Dell™ PowerEdge™ Expandable RAID Controller (PERC) 5 family of controllers, such as the configuration options, disk array performance, hardware specifications, redundant array of independent disks (RAID) management utilities, and operating system software drivers.
This section describes the hardware configuration features for the PERC 5 controllers.
Table 2-1 compares the configurations for the controllers.
|
NOTE: The maximum length of cable that you can use for Serial Attached SCSI (SAS) is 4 meters (13 feet) from port to port. This applies only to external cables. |
|
|
NOTE: The maximum array size is limited by the maximum number of drives per array (32) and the maximum number of spans per disk group (8), along with the size of the physical drives. This limits the number of spans in RAID 10 to eight, giving a total of 16 drives per virtual disk. |
The PERC 5 controllers recognize and use virtual disks created on existing PERC 5 controllers without risking data loss, corruption, redundancy, or configuration loss. Similarly, the virtual disks created on the controllers can be transferred to other PERC 5 controllers.
|
|
NOTE: For more information about compatibility, contact your Dell Technical Support Representative. |
The Self-Monitoring Analysis and Reporting Technology (SMART) feature monitors the internal performance of all motors, heads, and physical disk electronics to detect predictable physical disk failures. This feature helps monitor physical disk performance and reliability, and protects the data on the physical disk. When problems are detected on a physical disk, you can replace or repair the physical disk without losing any data.
SMART-compliant physical disks have attributes for which data (values) can be monitored to identify changes in values and determine whether the values are within threshold limits. Many mechanical and electrical failures display some degradation in performance before failure.
There are numerous factors that relate to predictable physical disk failures, such as a bearing failure, a broken read/write head, and changes in spin-up rate. In addition, there are factors related to read/write surface failure, such as seek error rate and excessive bad sectors.
|
|
NOTE: See www.t10.org for detailed information about Small Computer System Interface (SCSI) interface specifications and www.t13.org for Serial Attached ATA (SATA) interface specifications. |
Background Initialization (BGI) is a process to correct parity on the virtual disks. BGI is an automated check for media errors in which parity is created and written in the process. BGI does not run on RAID 0 virtual disks.
|
|
NOTE: You cannot permanently disable BGI. If you cancel BGI, it automatically restarts within five minutes. See Stopping Background Initialization for information on stopping BGI. |
The background initialization rate is controlled by the storage management software. You must stop an ongoing background initialization before you change the rate, or the rate change will not take effect. After you stop background initialization and change the rate, the rate change will take effect when the background initialization restarts automatically.
|
|
NOTE: Unlike initialization of virtual disks, background initialization does not clear data from the physical disks. |
|
|
NOTE: Consistency Check and Background Initialization perform the same function. The difference between them is that Background Initialization cannot be started manually, while Consistency Check can. |
The LED on the physical disk carrier indicates the state of each physical disk. For internal storage, see your system documentation for more information about the blink patterns.
For blink patterns on the Dell PowerVault™ MD1000, see the Dell PowerVault MD1000 Hardware Owner's Manual. For blink patterns on the Dell PowerVault MD3000, see the Dell PowerVault MD3000 Hardware Owner's Manual.
The PERC 5 controllers support moving physical disks from one cable connection or backplane slot to another on the same controller. The controllers automatically recognize the relocated physical disks and logically place them in the proper virtual disks that are part of the disk group. You can perform disk roaming only when the system is shut down.
Perform the following steps to use disk roaming.
The controller detects the RAID configuration from the configuration data on the physical disks.
The PERC 5 controllers support migration of virtual disks from one controller to another without taking the target controller offline. However, the source controller must be offline prior to performing the disk migration. The controller can import a virtual disk that is in optimal or degraded state. A virtual disk in offline state cannot be imported.
|
|
NOTE: The PERC 5 controllers are not backward compatible with previous SCSI PERC RAID controllers. |
When a controller detects a physical disk with a pre-existing configuration, it flags the physical disk as foreign and generates an alert indicating that a foreign disk was detected.
Dedicated hot spares are imported as global hot spares. The firmware generates an alert to indicate a change in hot spare configuration.
Perform the following steps to use disk migration.
The system with the target controller can be running while inserting the physical disks.
|
|
NOTE: Ensure the complete set of physical disks that form the virtual disk are migrated. |
An audible alarm is available on the PERC 5/E Adapter to alert you of key critical and warning events involving the virtual disk or physical disk problems. You can use the Basic Input/Output System (BIOS) Configuration Utility and management application to enable, disable, or silence the on-board alarm tone.
|
|
NOTE: See Audible Alarm Warnings for information about audible alarm codes. |
The Transportable Battery Backup Unit (TBBU) is a cache memory module with an integrated battery pack that enables you to transport the cache module with the battery into a new controller. The TBBU protects the integrity of the cached data on the PERC 5/E Adapter by providing backup power during a power outage.
The Battery Backup Unit (BBU) is a battery pack that protects the integrity of the cached data on the PERC 5/i Adapter and PERC 5/i Integrated controllers by providing backup power during a power outage. Unlike the TBBU, the BBU is not directly attached to the cache memory module and is therefore not transportable with the controller.
The TBBU and BBU offer an inexpensive way to protect the data on the memory module. The lithium battery provides a way to store more power in a smaller form factor than previous batteries.
See Transferring a TBBU Between Controllers for detailed procedures about handling controller cache in case of a controller failure.
The cache controller writes a block of data to cache memory, which is much faster than writing to the physical disk. The cache controller sends an acknowledgement of data transfer completion to the host system.
In write-through caching, the controller sends a data transfer completion signal to the host system when the disk subsystem has received all the data in a transaction. The controller then writes the cached data to the storage device when system activity is low or when the write buffer approaches capacity.
In write-back caching, the controller sends a data transfer completion signal to the host when the controller cache has received all the data in a transaction. The cached data is not written to the storage device.
The risk of using write-back cache is that the cached data can be lost if there is a power failure before it is written to the storage device. This risk is eliminated by using a battery backup unit on selected PERC 5 controllers. Refer to Table 2-1 for information on which controllers support a battery backup unit.
Write-back caching has a performance advantage over write-through caching.
|
|
NOTE: The default cache setting is write-back caching. |
|
|
NOTE: Certain data patterns and configurations perform better in a write-through cache policy. |
Firmware manages cache based on the condition of the battery. Learn Cycle is a battery calibration operation performed by the controller periodically (approximately every 3 months) to determine the condition of the battery.
Write-back caching is used under all conditions in which the battery is present and in good condition.
Write-through caching is used under all conditions in which the battery is missing or in a low-charge state. Low-charge state is when the battery is not capable of maintaining data for at least 24 hours in the case of a power loss.
Write-Back mode is available when the user selects Force WB with no battery. When Forced Write-Back mode is selected, the virtual disk is in Write-Back mode even if the battery is present and in good condition (healthy), or if a learn cycle is in process. Dell recommends that you use a power backup system when forcing Write-Back to ensure that there is no loss of data if the system suddenly loses power.
The time frame for completion of a learn cycle is a function of the battery charge capacity and the discharge/charge currents used. For PERC 5, the expected time frame for completion of a learn cycle is approximately seven hours and consists of the following parts:
Learn cycles shorten as the battery capacity derates over time.
|
|
NOTE: See the storage management application for additional information. |
During the discharge phase of a learn cycle, the PERC 5 battery charger is disabled. In this phase, the battery voltage is monitored through the smart battery bus (SMBus) using the battery gas-gauge. When the battery reaches the discharge capacity threshold (DCT), the dummy load is disabled and the battery charger is re-enabled. At this point, the battery charger detects that the battery is below the fast charge trigger voltage (FCTV) and initiates a fast-charge of the battery. The learn cycle completes once the battery fast charge is completed.
Table 2-2 lists the configuration features for the PERC 5 controllers.
Table 2-2. Features for RAID Configuration
Table 2-3 lists the features that provide fault tolerance to prevent data loss in case of a failed physical disk.
Table 2-3. Fault Tolerance Features
Hot swapping is the manual substitution of a replacement unit in a disk subsystem for a defective one, where the substitution can be performed while the subsystem is running (performing its normal functions).
|
|
NOTE: The backplane or enclosure must support hot swapping in order for the PERC 5 controllers to support hot swapping. |
|
|
NOTE: Ensure that SAS drives are replaced with SAS drives and SATA drives are replaced with SATA drives. |
|
|
NOTE: While swapping a disk, ensure that the new disk is of equal or greater capacity than the disk that is being replaced. |
The firmware automatically detects and rebuilds failed physical disks. Automatic rebuilds can be performed transparently with hot spares. If you have configured hot spares, the controllers automatically try to use them to rebuild failed physical disks.
The Patrol Read function is designed as a preventive measure that includes review of your system for possible physical disk errors that could lead to physical disk failure and damage data integrity. The Patrol Read operation can find and possibly resolve any potential problem with physical disks prior to host access. This can enhance overall system performance because error recovery during a normal I/O operation may not be necessary. You can use the storage management application to perform Patrol Read functions.
The following is an overview of Patrol Read behavior:
You can use the storage management application to select the Patrol Read options. Use Patrol Read options to set automatic or manual operation, or disable Patrol Read. The following sections describe Patrol Read functions and operations available in the Storage Management application.
|
|
NOTE: See the storage management application's documentation for more information about the Patrol Read configuration features available. |
The following describes the scheduling details for Patrol Read:
The behavior details of Patrol Read are as follows:
If any of the following conditions exist, then Patrol Read will not run on any of the affected disks: