Raid Array Data Recovery Glossary

Dave updated on Sep 22, 2022 to Knowledge Center

In the context of file storage, RAID or Redundant Array of Independent Disks is an essential term that signifies computer storage parts. This is responsible for copying data or transferring information across many drives.

While RAID helps with good-quality storage support, it has complexities like a complex structure, time-consuming steps, and unique RAID array terminology. As for the RAID array terms, here is a comprehensive glossary for you.


-- A --

·         Array- This refers to a group of two hard disk drives or more than the host device regarded as one device.

-- B --

·         Background Initialization- Alternatively called Immediate RAID Availability, this term represents the disk array initialization process that occurs entirely in the backdrop. After that, users can use the disk array in seconds instead of waiting for hours.

·         Benchmarks- This term refers to the criteria list of conditions that becomes the measurement scale for any system or product. The PC system trade magazine labs take the time to evaluate frequently and draw comparisons based on the benchmarks about new computer devices or systems. They study specific criteria, like context-based events, user interactions, or application program types. It is possible to design a program that supports specific app or OS-related measurements.

·         Berkeley RAID Levels- Experts like David Patterson, Randy Katz, and Garth Gibson mentioned specific mapping and protection techniques for disk arrays while researching input/output (I/O) systems at Berkeley University in California. They compiled their findings into six levels, from RAID 0 to 5.

·         Bridge RAID Controller- This is also known as the External RAID controller. This device comes in a single ID form visible within SAN (Storage Area Network). It connects with multiple devices for controlling varying external RAID subsystems.

-- C --

·         Cache- This quick and temporary storage section keeps specific data from another slower storage device. Users can access it faster, and it exists transparently in the system.

·         Cache Flush- The operation for coding Write-Back Cache's unwritten blocks towards the target disk for powering the system down is called caching.

·         Cache Line Size- This indicates the data amount one has to write or read at a time. They appear in conjunction with the stripe size.

·         Conservative Cache- This is the operating mode of the system drivers after the Write-Back Caching configuration that works like it was configured for the Write-Through Caching.

·         Consistency Check- The process of checking redundant data integrity, specifically in RADI 0+1 and RAID 1, is referred to as a consistency check. For RAID 3 or 5, this checking involves comparing the written parity and disk parity.

·         Controller- This module or adapter card interprets signals and controls them between the peripheral device and host.

-- D --

·         Degraded Mode- This is the RAID mode that works in place of a failed component drive.

·         Disk- This is a non-volatile and rewritable device for data storage that includes rotating magnetic plus optical disks. It also includes electronic storage types and SSDs.

·         Disk Array- A group of disks from more than one easy-to-access disk system, also referred to as RAID. They are commonly used with workstations and desktops, mainly active on servers.

·         Disk Drive- This type of device stores digitally available data electronically.

·         Disk Failure Detection- RAID controllers easily analyze and find SCSI-based disk failures. They run a monitoring process for controllers to find elapsed command periods. Time-outs result in a retrial of command and disk resetting.

·         Disk Traveling, Drive Traveling- These terms refer to the process when drivers exist in another order than their original style. Here, disk traveling is possible even with failed drive status.

·         Disk System- This type of storage system can hold disks only.

·         Drive Groups, Drive Packs- These RAID array terms refer to grouped disk drives that are identical that are attached. They work as one unit. In the case of physical devices, the drive groups are called 'drive packs'.

·         Dual Active- There are two components within the failure-tolerant storage system, like a storage controller pair. These work together on one or more tasks while functioning regularly. The second component takes over the load if one fails.

-- E --

·         ECC- Known as Error Correction Code, this is a redundant data generation method for finding and correcting errors within transmitted/stored data.

·         Embedded Storage Controller- This competent storage controller sits atop the host system's housing. It fixes directly into the memory bus of the host without any I/O bus/adapter.

·         External RAID Controller- This refers to the RAID controllers within their personalized enclosures instead of the server or PC system. They are also known as Bridge RAID controllers.

-- F --

·         Failback- This indicates the load share restoration of a failed component in a computer system to another replacement component.

·         Failover- The systems that handle system failure can use this operation mode in case of component failure in one system and when the redundant component begins its operations.

·         Failover Port- This fiber channel-based port can assure the I/O requests to a failed loop port. This stays active during normal operations, sometimes dormant. These ports use the same node and loop ID.

·         Failure- The term in the RAID array glossary indicates the physical issues or defects in hardware types that are easy to detect and need replacement.

·         Fault Tolerance, Failure Tolerance- The term refers to a system's ability to continue with normal functioning after a component failure. Fault-tolerant systems need redundancy within adapters, power supplies, disk drives, cabling, and controllers.

·         Fibre Channel- This technology is helpful for moving data from computer to computer at 2 Gbps or lower data rate. This is suitable for computer systems within a shared ecosystem with connected servers. It also works with interconnected drives or storage controllers. It has better flexibility than SCSI and will work as a transmission replacement.

·         Flash ROM- In the RAID array glossary, this signifies the adapter memory within the software that developers can re-code directly from the board.

-- G --

·         Global Spare- An additional disk drive (physical type) for an array that works as the hot spare. This can work in place of a failed drive automatically.

-- H --

·         Host- This refers to the primary computer system with which the RAID array is attached.

·         Hot Spare- Among the RAID array terms, hot spare refers to a hard drive type synced with the array but not actively operating. If another drive within the array fails, users can opt for it as a spare.

·         Hot Standby- This unused part of the fault-tolerant storage system is ready to go active at any time. However, it is a substitute for another primary component. It will only work when the first one expires. 

-- I --

·         Immediate RAID Availability- Another term for Background Initialization (see under section B).

·         In-Line Terminator- This is a type of plugin at a SCSI cable's end for starting active termination processes in case of the lack of any built-in termination.

·         Interface- This is the type of software/hardware protocol that handles data exchanges between the host and hard disk drives. Typical examples for smaller systems include SCSI and ATA.

·         Internal RAID Controller- This refers to the controller-based circuit board within the server or computer, mainly on the bus like the PCI bus.

-- J --

·         JBOD- These are multiple disk drives, known as "Just A Bunch of Disks," all kept within one enclosed space. They function without RAID technology, and this mode is available in Mylex RAID controllers.

-- K --

·         Kilobyte (KB)- This is a type of measurement.

-- L --

·         Latency- This term refers to the time between requesting I/O and the final request completion period. It can also signify rotation latency, which indicates the interval between a seek completion and the first data block arrival for the disk's read and write head.

·         Logical Drive- These devices work with the OS and are presented as a free disk drive with high capacity.

-- M --

·         Mirrored Cache- This type of cache memory duplicates another controller possesses. So, when the original controller fails, it can work as a backup.

·         Mirrored Hard Drive- The host has two hard disk drive types as a single unit. Data is synced on both drives simultaneously for quick dependency on one in case the other gets damaged.

·         Mirroring- This term refers to the duplication of information from the source drive to another drive.

-- O --

·         Overwrite- The process of coding on top of other existing data.

-- P --

·         Parity- This is a type of information security measure that RAID 5 uses for recreating failed drive data within the array.

·         Parity Check- This is the process of offering full data redundancy while the system needs only one section of the mirroring capacity for storage.

·         Partitioning- The total disk/disk array storage capacity comes up within the operating environment as multiple virtual disks handle the extra underlying capacity.

·         PCI Hot Plug- This feature allows the PCB to not power down the system before allowing the user to add or configure new changes. 

-- Q --

·         Queue- Multiple functions, tasks, or components in a waiting line.

-- R --

·         RAID- refers to multiple disks operating together in one group for improved storage limit.

·         RAID Advisory Board (RAB)- A group of companies working together for RAID storage system standardization.

·         RAID Controller- These low-cost RAID controllers include SCSI channels within their motherboards.

·  RAID Migration- This feature within RAID subsystems for transitioning between RAID levels without system power down.

·  Read-Ahead Cache- This is a caching method where the PC anticipates the data and keeps the cache on hold for quick request resolution.

·  Recovery- The process of getting data from other drives to put together after a failed disk condition.

· Redundancy- This refers to adding multiple components of one type within the computer system, more than it actually needs at the time.

· Rotated XOR Redundancy- This method offers total data redundancy in exchange for only a select amount of storage capacity for mirroring.

-- S --

·  SAF-TE- SCSI Accessed Fault-Tolerant Enclosure is a type of standardization testing for comprehensive testing and evaluation of various elements like cooling system, disk drives, and power supplies of LAN servers/storage subsystems.

·  Sector- This refers to the place of a fixed-block type of architecture disk where the data that is at the risk of deletion is saved.

·  Segment Size- Similar to Cache Line Size (see section C).

· Sequential I/O- This refers to a form of reading and writing operation for assessing entire data blocks in sequential order.

· SES- SCSI Enclosure Services is a type of SCSI standard for accessing the services related to enclosures of two or more SCSI-based devices.

·  Session- This refers to the interval of time that occurs between two consecutive system-based shutdowns, like after intentional power off or hardware reset.

·  SMART- Abbreviation for "Self-Monitoring Analysis and Reporting Technology"; this is an indicator for predicting the reliability of SCSI and ATA/IDE hard disk drives. This is an industry-standard.

· Spanning- The procedure for configuring sections of many drive packs together or whole drive packs in one go. So, the data processing volume recorded here is more than single drives.

·   Standard Disk Drive- These are complex disk drive types with IDE or SCSI, among other types, and are linked with the host computer via a standard-level disk controller.

· Standby Replacement of Disks- Related to Hot Spare (see section H).

·  Storage Device- This is a collective word describing tape transports, disks, and other types of mechanisms that handle non-volatile data storage.

·  Stripe Width- The total number of striped SCSI drives that existed in one drive group constitutes the stripe width.

·  Striping- The spread of information evenly between multiple drives to improve the overall performance.

·  Subsystem Storage- The term refers to a total collection of multiple disks that offer storage space for extra data to system users.

-- T --

·  Terminator- It sits at the end of any SCSI bus.

·  Termination- This is the process of aligning a bus' transmission impedance to remove any existing signal reflections the bus ends make.

· Throughput- The throughput signifies the total I/O requests completed per unit of time, like seconds.

·  TPC-C, TPM-C- Known as Transaction Processing Performance Council, the standardization organization calculates the throughput transactions within systems. The TPM-C benchmark refers to performance and price metrics, while TPC-C focuses on the transaction speed of new orders.

-- U --

· Unformatted Capacity- This error type shows that the logical drive did not undergo any formatting even after hours of expansion attempts and steps.

· Upgrade- The process of upscaling to improved or bigger-sized hard disks.

-- V --

· Volume- Multiple hard drives in the RAID levels, like RAID 0, create one virtual drive for the operating system in the host computer, known as volume.

-- W --

·  Write-Back Cache- This is a caching method where the completed signal's written commands are provided to the host OS instantly after the cache gets the data request.

·  Write-Through Cache- This type of caching strategy happens where the SCSI drive records the written data before the host OS gets the completion status. Comparatively, it is more secure than Write-Back Cache since a power failure will not cause data loss.

-- X --

·  XOR- This computer coding language function, called Exclusive "Or" i.e., XOR, develops parity within different RAID systems.

-- Z --

· Zoned Recording- This indicates data recording in specific zones after storage array configuration.


Overall, there are many RAID array terms to know and understand that you should start reading up. However, for an easier experience without such complex storage optimization factors, you can use EaseUS Todo Backup instead.

Was This Page Helpful?