RAID Controller Functions

In the context of RAID terms are used and functions described which are explained in detail below.

Embedded RAID / Host RAID

"Embedded RAID" means hardware (ASIC) is provided on the RAID controller which relieves the system CPU (host) of RAID controller functions. As a result the server can devote itself to processing its core applications and the overall performance is improved. If this hardware support is not available, the term "host RAID" is used.

Drive Usage

For simplicity's sake the different RAID types under Supported RAID Levels all use complete disk drives of the same size. In fact the usable capacity of each drive is limited by the disk drive with the lowest capacity if drives with different capacities are used.

If, for example, a RAID-1 is created from a 160-Gbyte and an 80-Gbyte drive, only half the capacity of the larger disk can be used, thus limiting it to 80-Gbyte. Furthermore, a small part is removed from each drive for the so-called RAID Signature.

RAID Signature

RAID controllers use a small segment at the beginning or end of each connected drive to store information on the drives and arrays attached to the controller. This segment is also referred to as RAID Signature and is not available for general use for saving user data.

Migration

Some RAID controllers support the modification of existing logical drives through expansion options, migration of one RAID type to another, and modification of the stripe size. The migration options depend on the RAID controller used.

Further information is provided under Modifying Logical Drives.

Online Capacity Expansion

Most operating systems of today support Online Capacity Expansion (OCE). OCE means that the additional capacity after a logical drive has been expanded can be used without rebooting the system. Details on the additional storage capacity are provided in the documentation for your operating system.

Cabinets

RAID controllers also support external drive enclosures which use SES or SAF-TE enclosure administration hardware. This extended hardware support enables additional administrative information for the enclosure, e.g. fan speed, temperature and voltage. Such enclosures generally offer further properties, e.g. hot swap.

Hot Swap

Either through the use of SATA technology or the above-mentioned drive enclosure RAID controllers support so called hot swap, i.e. disk drives can be replaced during ongoing operation without the system having to be rebooted.

Note: Hot swapping of hard disks is possible only if a disk was placed Offline beforehand.

Hot Spare

A hot spare is a physical drive which is available in a redundant logical drive as a replacement for a failed disk. If a drive fails the hot spare replaces it and the logical drive is recreated. The data is then reconstructed on this new disk during ongoing operation. Until reconstruction has been completed the access to the data takes a little longer but is possible at any time.

RAID controllers support the following hot spare types:

Note: Some RAID controllers automatically assign newly added and unused drives to the global hot spares.

Consistency Check/Make Data Consistent (MDC)

The consistency check operation verifies correctness of the data in logical drives that use RAID levels 1, 5, 6, 10, 50, and 60. (RAID-0 does not provide data redundancy). For example, in a system with parity, checking consistency means computing the data on one disk and comparing the results to the contents of the parity disk.

Make data consistent (MDC) does not only verify the correctness of data but also attempts to correct inconsistent data automatically.

Note: It is recommended that you perform a consistency check at least once a month.

Copyback

The copyback feature allows you to copy data from a source disk of a logical drive to a destination disk that is not a part of the logical drive. Copyback is often used to create or restore a specific physical configuration for an array (for example, a specific arrangement of array members on the device I/O buses). Copyback can be run automatically or manually.

Typically, when a disk fails or is expected to fail, the data is rebuilt on a hot spare. The failed disk is replaced with a new disk. Then the data is copied from the hot spare to the new disk, and the hot spare reverts from a rebuild disk to its original hot spare status. The copyback operation runs as a background activity, and the logical drive is still available online to the host.

Copyback is also initiated when the first Self-Monitoring Analysis and Reporting Technology (SMART) error occurs on a disk that is part of a logical drive. The destination disk is a hot spare that qualifies as a rebuild disk. The disk with the SMART error is marked as failed only after the successful completion of the copyback. This avoids putting the array in degraded status.

Background Initialization (BGI)

Background initialization is a consistency check that is forced when you create a virtual drive. This is an automatic operation that starts 5 minutes after you create the virtual drive.

Background initialization is a check for media errors on the disks. It ensures that striped data segments are the same on all disks in a drive group. The default and recommended background initialization rate is 30 percent. Before you change the rebuild rate, you must stop the background initialization or the rate change will not affect the background initialization rate.

Patrol Read

Patrol read involves the review of your system for possible disk errors that could lead to disk failure and then action to correct errors. The goal is to protect data integrity by detecting disk failure before the failure can damage data. The corrective actions depend on the array configuration and the type of errors.

Patrol read starts only when the controller is idle for a defined period of time and no other background tasks are active, though it can continue to run during heavy I/O processes.

MegaRAID® CacheCade™ Pro 2.0

MegaRAID CacheCade Pro 2.0 read/write software eliminates the need for manually configured hybrid arrays by intelligently and dynamically managing frequently accessed data and copying it from HDD volumes to a higher performance layer of SSD cache. Copying the most accessed data (‘hot spot’) to flash cache relieves the primary HDD array from time-consuming transactions which allows for more efficient hard disk operation, reduced latency, and accelerated read and write speeds. This provides significant improvements to overall system performance – two to twelve times that of HDD–only configurations – for a wide variety of server applications including web, file, online transaction processing (OLTP) database, data mining and other transaction-intensive applications.

MegaRAID® FastPath™

MegaRAID FastPath software is a high-performance IO accelerator for SSD arrays connected to a MegaRAID controller card. This advanced software is an optimized version of MegaRAID technology that can dramatically boost storage subsystem and overall application performance – particularly those that demonstrate high random read/write operation workloads – when deployed with a 6Gb/s MegaRAID SATA+SAS controller connected to SSDs.

Application workloads that will benefit most from MegaRAID FastPath software with SSD volumes are those with small and random IO patterns requiring high transactional throughput, such as OLTP.