In the context of file storage, RAID is a system that copies data across multiple drives, ensuring that if one drive fails, the data is still accessible from the other drives.
While RAID helps with good-Quality storage support can be complex, with intricate structures, time-consuming steps, and specialized RAID array terminology that can be overwhelming. To help navigate this, here is a comprehensive glossary of RAID array terms.
· ArrayA group of two or more hard disk drives that are considered as one device by the host, is known as a RAID.
· Background Initialization- The term Immediate RAID Availability refers to the disk array initialization process that happens in the background, allowing users to access the disk array in seconds, rather than waiting for hours.
· BenchmarksThe benchmark is a set of criteria used to measure the performance of a system or product. PC system trade magazines like Labs regularly evaluate and compare new computer devices or systems based on these benchmarks, which consider factors such as context-based events, user interactions, and application program types. This allows for the design of programs that support specific app or OS-related measurements.
· Berkeley RAID LevelsExperts like David Patterson, Randy Katz, and Garth Gibson researched I/O systems at Berkeley University and identified six levels of disk array mapping and protection techniques, from RAID 0 to 5.
· Bridge RAID Controller- The External RAID controller is also known as a single ID device visible within a SAN. It connects with multiple devices to control external RAID subsystems.
· CacheThis section temporarily stores specific data from a slower storage device, allowing users to access it faster, and operates 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, appearing in conjunction with the stripe size.
· Conservative CacheThe system drivers operate in a mode that mimics Write-Through Caching, even after Write-Back Caching is configured.
· Consistency Check- In RAID systems, a consistency check is performed to verify the integrity of data by comparing the written parity with the disk parity, especially in RAID 3 or 5 configurations. This process ensures that the data is consistent and has not been corrupted.
· Controller- This module or adapter card interprets signals and controls them between the peripheral device and host, facilitating communication and data exchange.
· Degraded Mode- In the event of a failed component drive, the RAID mode can seamlessly take its place, ensuring uninterrupted operation and maintaining data integrity.
· Disk- A hybrid data storage device combines rotating magnetic and optical disks with electronic storage types, including solid-state drives (SSDs), offering a non-volatile and rewritable storage solution.
· Disk Array- A group of disks from more than one disk system, also referred to as RAID, are commonly used with workstations and desktops, mainly active on servers.
· Disk Drive- This type of device stores digitally available data electronically.
· Disk Failure DetectionRAID controllers can easily detect and identify SCSI-based disk failures by running a monitoring process that tracks elapsed command periods. If a time-out occurs, the controller will retry the command and reset the disk.
· Disk Traveling, Drive Traveling- These terms refer to the process when drivers exist in an order other than their original style, allowing disk traveling even with a 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 and attached, working as one unit, and are called 'drive packs' when referring to physical devices.
· Dual Active- The failure-tolerant storage system consists of a pair of storage controllers that work together to perform tasks normally, and a secondary component that takes over the load if one of the primary controllers fails.
· ECC- Error Correction Code is a method of generating redundant data to detect and correct errors in transmitted or stored data.
· Embedded Storage ControllerThe storage controller is a competent component that sits on top of the host system's housing, directly fixing into the memory bus of the host without the need for an I/O bus or adapter.
· External RAID ControllerThis refers to the RAID controllers within their personalized enclosures, also known as Bridge RAID controllers, rather than the server or PC system itself.
· FailbackThis indicates the load share restoration of a failed component in a computer system to another replacement component.
· FailoverIn the event of a system component failure, a system can switch to a backup operation mode, seamlessly transitioning to a redundant component when it becomes available.
· Failover Port- A fiber channel-based port can maintain active I/O requests to a failed loop port, staying dormant during normal operations. It shares the same node and loop ID as the original port.
· Failure- The term refers to physical issues or defects in hardware that are easily detectable and require replacement.
· Fault Tolerance, Failure Tolerance- A fault-tolerant system is one that can continue to function normally even if a component fails, requiring redundancy in adapters, power supplies, disk drives, cabling, and controllers.
· Fibre Channel- This technology allows for fast data transfer between computers at speeds of 2 Gbps or lower, making it suitable for shared computer systems and interconnected drives or storage controllers, offering better flexibility than SCSI and serving 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.
· Global SpareAn additional disk drive that can automatically replace a failed drive in an array, functioning as a hot spare.
· HostThis refers to the primary computer system with which the RAID array is attached.
· Hot SpareIn RAID, a hot spare is a non-active hard drive that mirrors the data of the other drives in the array, and can be automatically swapped in if one of the other drives fails.
· Hot StandbyThe redundant storage system is on standby, waiting to replace the primary component when it fails.
· Immediate RAID Availability- Another term for Background Initialization (see under section B).
· In-Line TerminatorA type of SCSI cable plug that initiates active termination when no built-in termination is present.
· InterfaceThis software/hardware protocol handles data exchanges between the host and hard disk drives, with examples including SCSI and ATA for smaller systems.
· Internal RAID ControllerThis refers to the controller-based circuit board within the server or computer, primarily on the bus like the PCI bus.
· JBOD- The "Just A Bunch of Disks" (JBOD) mode allows multiple disk drives to be housed in a single enclosure, without the use of RAID technology. This mode is typically available in Mylex RAID controllers.
· Kilobyte (KB)- This is a type of measurement.
· LatencyThis term refers to the time between requesting I/O and the final request completion period, or the interval between a seek completion and the first data block arrival for the disk's read and write head.
· Logical DriveThese devices work with the operating system and are presented as a free disk drive with high capacity.
· Mirrored Cache- This type of cache memory serves as a backup controller, duplicating the original controller's function, so that if the original one fails, the duplicate can take over.
· Mirrored Hard Drive- The host has two hard disk drives that work together as one unit, syncing data on both drives at the same time, so that if one drive is damaged, the other can take over quickly.
· Mirroring- This term refers to the process of copying information from one drive to another.
· OverwriteThe process of coding on top of other existing data is referred to as data overlay or data integration. This involves combining new data with existing data to create a new dataset, often to gain new insights or to enhance the existing data.
· ParityRAID 5 uses parity data to recreate failed drive data within the array.
· Parity Check- The process of offering full data redundancy while the system needs only one section of the mirroring capacity for storage is called RAID. This allows for the system to have a backup of its data, but only requires one part of the mirrored data to be stored, making it a space-efficient solution.
· Partitioning- The total storage capacity of the disk/disk array is increased within the operating environment as multiple virtual disks utilize the extra underlying capacity.
· PCI Hot Plug- This feature enables the system to remain powered on, allowing users to make and configure changes without the system shutting down.
· Queue- Multiple functions, tasks, or components in a waiting line.
· RAIDA RAID (Redundant Array of Independent Disks) is a storage technology that combines multiple disks into a single logical unit to provide improved storage capacity, performance, and redundancy. By grouping multiple disks together, a RAID setup can offer increased storage capacity, faster data transfer rates, and protection against data loss in case of a disk failure.
· RAID Advisory Board (RAB)- A group of companies is working together to standardize RAID storage systems.
· RAID ControllerThese low-cost RAID controllers include SCSI channels within their motherboards.
· RAID Migration- The feature within RAID subsystems allows for transitioning between different RAID levels without requiring the system to be powered down.
· Read-Ahead CacheThis is a caching method where the PC anticipates the data and keeps the cache on hold for quick request resolution.
· RecoveryAfter a failed disk condition, the process of retrieving data from other drives to compile and put together is necessary.
· RedundancyThis 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.
· SAF-TE- SCSI Accessed Fault-Tolerant Enclosure is a standardization testing method that comprehensively evaluates and tests various components of LAN servers/storage subsystems, including the cooling system, disk drives, and power supplies, to ensure their reliability and performance.
· SectorThis refers to the place on a fixed-block type of architecture disk where data at risk of deletion is saved.
· Segment Size- Similar to Cache Line Size (see section C).
· Sequential I/OThis refers to a form of reading and writing operation for assessing entire data blocks in sequential order.
· SESSCSI Enclosure Services (SES) is a type of SCSI standard that provides access to the services related to enclosures of two or more SCSI-based devices.
· SessionThis refers to the time between two consecutive system-based shutdowns, such as after a power off or hardware reset.
· SMART- SMART is an industry-standard indicator for predicting the reliability of SCSI and ATA/IDE hard disk drives.
· Spanning- Configuring sections of many drive packs together or whole drive packs in one go requires a procedure that allows for data processing volumes to be recorded that are more than what single drives can handle.
· Standard Disk DriveThese are complex disk drive types with IDE or SCSI interfaces, among other types, and are connected to the host computer via a standard-level disk controller.
· Standby Replacement of Disks- Related to Hot Spare (see section H).
· Storage DeviceThis is a collective term that encompasses tape transports, disks, and other mechanisms for handling non-volatile data storage.
· Stripe Width- The total number of striped SCSI drives in one drive group is known as the stripe width.
· StripingThe goal of distributing information evenly across multiple drives is to enhance overall system performance. By spreading data across multiple drives, the load on each individual drive is reduced, allowing for faster access and processing times.
· Subsystem StorageThe term refers to a collection of multiple disks that provide additional storage space for system users.
· Terminator- It sits at the end of any SCSI bus.
· TerminationThis process involves aligning a bus' transmission impedance to remove existing signal reflections caused by the bus ends.
· ThroughputThe throughput is a measure of the total I/O requests completed per unit of time, typically expressed in seconds.
· TPC-C, TPM-C- The Transaction Processing Performance Council (TPC) measures the throughput of transactions within systems, calculating performance and price metrics through the TPM-C benchmark, and focusing on transaction speed through TPC-C, which specifically evaluates the speed of new orders.
· Unformatted CapacityThe error indicates that the logical drive, despite hours of expansion attempts and steps, did not undergo any formatting, suggesting a potential issue with the drive's formatting or expansion process.
· UpgradeThe process of upgrading to better or larger-sized hard disk drives.
· VolumeMultiple hard drives in RAID levels, such as RAID 0, combine to create a single virtual drive, referred to as a volume, which is utilized by the operating system in the host computer.
· Write-Back CacheThis caching method involves providing the host OS with the completed signal's written commands immediately after the cache receives a data request, allowing for instant data access.
· Write-Through CacheIn Write-Through Cache, the SCSI drive records written data before the host OS receives completion status, making it more secure than Write-Back Cache, as a power failure won't result in data loss.
· XOR- The Exclusive "Or" (XOR) function in computer coding is used in RAID systems to develop parity, which ensures data integrity and redundancy.
· Zoned Recording- This indicates data recording in specific zones after storage array configuration. The data is recorded in the specified zones after the storage array has been configured.
There are many RAID array terms to know and understand, but for an easier experience, you can use Qiling Backup instead.