hesela.devCorpus updates
Hesela team · RSS feed · Observatory updates
New analysis RSS · Documented corrections RSS
Site notes
: added corpus documentation, citation and correction links, content-type search filters, and this revision index.
Dated glossary revisions
Each entry uses the record's explicit editorial update date, not a build, download or source-access date. This is an index of latest recorded revisions, not a complete change history. Undated records and datasets are not assigned invented dates.
- · Adaptive replica selection
Adaptive replica selection routes a request among eligible data replicas using changing performance or load information rather than a fixed choice.
- · Annualized failure rate (AFR)
Annualized failure rate (AFR) expresses observed failures relative to drive-time exposure, scaled to a year, and is not inherently a one-year failure probability.
- · Average precision
Non-interpolated average precision summarizes ranked prediction scores by weighting precision at each threshold by the corresponding increase in recall.
- · Backup verification
Backup verification checks stored backup material against specified integrity and completeness rules.
- · Belady anomaly
A Belady anomaly occurs when a replacement policy produces more misses with a larger cache than with a smaller cache on the same reference string and initial-state convention.
- · Binomial proportion
A binomial proportion is an outcome count divided by a fixed number of independent binary trials, used to estimate their common outcome probability.
- · Bit-patterned media
Bit-patterned media is a magnetic recording medium that stores individual bits on lithographically defined, separated magnetic islands.
- · Blocked Bloom filter
A Bloom-filter layout that confines the bit probes for a key to one selected block to improve memory locality.
- · Bloom filter
A compact approximate-membership structure that can reject absent keys but may report a false positive, without false negatives for correctly inserted keys in an intact standard filter.
- · Brier score
For a binary event, the Brier score is the mean squared difference between each forecast probability and its observed zero-or-one outcome.
- · Buffered I/O
Buffered I/O is file access through an operating-system cache, allowing transfers between an application and storage to be decoupled.
- · Cache inclusion property
The cache inclusion property requires the resident set at a smaller capacity to be contained in that at a larger capacity after each shared request prefix under the same replacement policy.
- · Cache pollution
Cache pollution occurs when low-reuse entries displace entries that would otherwise serve useful cache hits.
- · Charge trap flash
A flash cell design that stores charge in localized traps within an insulating material rather than on a conductive floating gate.
- · CLOCK cache replacement
CLOCK is a cache replacement policy that gives referenced entries another chance by clearing their reference state during an eviction scan.
- · close
close is an operation that releases a file descriptor so its number can be allocated again.
- · Confidence interval
A confidence interval is a set of parameter values produced by a procedure with a stated repeated-sampling coverage under specified model and sampling assumptions.
- · Counting Bloom filter
A counting Bloom filter tracks approximate set membership with counters that support insertions and deletions through matched increments and decrements.
- · Crash-corruption ambiguity
Crash-corruption ambiguity occurs when the same observed stored state is consistent with both an interrupted update and corruption after persistence.
- · Cuckoo filter
A cuckoo filter is an approximate-membership structure that stores short fingerprints in candidate buckets and supports insertion, lookup and deletion subject to its update contract.
- · Data scrubbing
Data scrubbing is proactive verification of stored data to discover errors before ordinary access requires the affected content.
- · Drive-day
A drive-day is one drive observed in operation for one day, used as a unit of fleet exposure.
- · Expected calibration error
Expected calibration error is a summary of disagreement between predicted probabilities and conditional outcome frequencies, commonly estimated by bin-weighted gaps on an evaluation sample.
- · F1 score
F1 score is the harmonic mean of precision and recall for a specified positive class and decision threshold.
- · Failure detector
A failure detector is a component that uses observations to report whether another process or endpoint is suspected of being unavailable.
- · File descriptor
A file descriptor is a nonnegative integer used by a process to refer to an open file description for operations such as reading, writing or closing.
- · File truncation
File truncation changes a regular file to a specified byte length, discarding data beyond a shorter end or extending it when supported.
- · First-in first-out cache replacement
First-in first-out cache replacement evicts the resident entry that was inserted earliest, without refreshing its queue position on a hit.
- · fsync
fsync is a file-synchronization operation that requests persistence of a file's modified data and associated metadata before reporting completion.
- · Full-page write
A PostgreSQL full-page write records a recoverable page image in WAL to support recovery from an incomplete data-page write.
- · Hard link
A hard link is a directory entry naming an existing file's inode, so multiple hard links refer to the same file rather than to separate copies.
- · Heat-assisted magnetic recording (HAMR)
Heat-assisted magnetic recording uses localized optical heating to enable magnetic writing of high-anisotropy HDD media.
- · Hedged request
A hedged request is a redundant copy dispatched after an initial request remains unfinished for a chosen delay, allowing an eligible response to complete the operation.
- · Inspection paradox
The inspection paradox is the length bias by which sampling a time point overrepresents long intervals compared with sampling intervals equally.
- · Interarrival time
An interarrival time is the elapsed interval between consecutive events in a specified event stream.
- · io_uring
io_uring is a Linux asynchronous I/O interface that exchanges requests and completion results through shared submission and completion rings.
- · io_uring cancellation
io_uring cancellation is a request to stop matching outstanding operations, with a result separate from each target operation's completion.
- · io_uring linked requests
io_uring linked requests are a chain of submission entries whose dependencies delay each successor until its predecessor completes.
- · io_uring linked timeout
An io_uring linked timeout is a timer request attached to an operation that can trigger cancellation when its deadline expires.
- · iSCSI
iSCSI is a storage transport protocol that carries SCSI commands, data and responses between initiators and targets over TCP/IP networks.
- · iSCSI Command Sequence Number
An iSCSI Command Sequence Number (CmdSN) identifies a command's position in the session-wide delivery sequence.
- · iSCSI Connection
An iSCSI connection is a TCP connection carrying iSCSI protocol messages between an initiator and a target within an iSCSI session.
- · iSCSI Error Recovery Level
The iSCSI Error Recovery Level is the session-wide negotiated value from 0 to 2 that selects the protocol recovery capabilities available to an initiator and target.
- · iSCSI Network Portal
An iSCSI network portal is a TCP/IP access point used by an iSCSI node, identified by an IP address at an initiator and by an IP address plus listening TCP port at a target.
- · iSCSI Node Name
An iSCSI node name is the persistent, location-independent identifier of an iSCSI initiator or target, distinct from the IP addresses through which that node is reached.
- · iSCSI Portal Group
An iSCSI portal group is a node-scoped set of network portals capable of coordinating the connections of a common iSCSI session.
- · iSCSI Session
An iSCSI session is a protocol association between an initiator and a target that groups TCP connections and, for a normal session, represents a SCSI I_T nexus.
- · iSCSI Status Sequence Number
An iSCSI Status Sequence Number (StatSN) numbers target responses within one connection for status acknowledgement and recovery.
- · Journal checksum
A journal checksum is an integrity value computed over specified journal records or payloads to detect changes within that coverage.
- · Lateral charge spreading
Lateral charge spreading is migration of stored charge along a shared trapping layer between neighboring cells in charge-trap NAND.
- · Least recently used replacement
Least recently used replacement evicts the resident entry whose most recent access is oldest.
- · Magnetic anisotropy
Magnetic anisotropy is the dependence of magnetic energy or response on the direction of magnetization.
- · Magnetic recording trilemma
The magnetic recording trilemma is the coupled constraint of signal-to-noise ratio, thermal stability and writability as recording density increases.
- · Magnetic thermal stability factor
The magnetic thermal stability factor is the ratio of a magnetization-reversal energy barrier to thermal energy at a specified absolute temperature.
- · Massive array of idle disks
A massive array of idle disks is a storage architecture that keeps selected disks spun down between accesses to reduce energy use.
- · Matthews correlation coefficient
For binary classification, the Matthews correlation coefficient is the Pearson correlation between predicted class labels and observed outcomes, calculated from all four confusion-matrix counts.
- · Mean time to detection
Mean time to detection is the arithmetic mean of elapsed times between event occurrence and detection for a specified event population.
- · Memory-mapped file
A memory-mapped file exposes a file region through a process's virtual address space so its contents can be accessed with memory loads and stores.
- · mincore
mincore is a system call that reports a page-residency vector for a mapped virtual-address range.
- · MLC SSD
An MLC SSD uses NAND flash that, in the two-bit sense of MLC, encodes two bits per cell using four threshold-voltage states.
- · msync
msync is an operation that synchronizes a range of a shared file-backed memory mapping with its underlying file, with MS_SYNC waiting for completion.
- · Nearline HDD
A nearline HDD is a vendor category of capacity-oriented enterprise hard disk drives commonly used for bulk online storage, distinct from nearline as a storage-availability class.
- · Neel-Arrhenius model
The Neel-Arrhenius model describes thermally activated magnetic reversal with an exponential dependence on the ratio of an energy barrier to thermal energy.
- · NFS
Network File System is a family of protocols for accessing files and directories on a remote server over a network.
- · NFS COMMIT
NFSv3 COMMIT requests stable storage of previously buffered writes and associated metadata for a specified file range.
- · NFS write gathering
NFS write gathering combines related server writes to share persistence work before sending their replies.
- · NVMe
NVMe is a family of storage command and transport specifications for accessing non-volatile memory over PCIe and fabrics.
- · NVMe discovery controller
An NVMe discovery controller supplies information for locating NVM subsystems or other discovery services without exposing namespaces or implementing I/O queues.
- · NVMe over Fabrics
An extension that carries NVMe commands across RDMA, Fibre Channel, or TCP networks to remote namespaces.
- · NVMe Physical Region Page
An NVMe Physical Region Page entry identifies a memory page or page portion used to transfer a command's data buffer.
- · NVMe Scatter Gather List
An NVMe Scatter Gather List describes a command's data transfer using typed descriptors for buffers or descriptor segments.
- · NVMe/FC
NVMe/FC is the transport binding that carries NVMe capsules and data over Fibre Channel.
- · NVMe/RDMA
NVMe/RDMA is an NVMe transport binding that uses reliable RDMA queue pairs for capsule delivery and direct data placement.
- · NVMe/TCP
NVMe/TCP is an NVMe transport binding that carries commands, responses and data as protocol data units over TCP connections.
- · NVMe/TCP digest
An NVMe/TCP digest is an optional CRC32C error-detection value protecting a protocol data unit's header or data after negotiation at connection establishment.
- · O_APPEND
O_APPEND is a file-status flag that combines positioning at the current end of a file with each write as one atomic operation.
- · O_DSYNC
O_DSYNC is an open-file status flag requiring write operations to complete with synchronized I/O data integrity, covering data and the metadata needed to retrieve it.
- · O_TMPFILE
O_TMPFILE is a Linux open flag that creates a regular file without a directory entry in a specified directory's filesystem.
- · Open file description
An open file description is a system-wide open-file object referenced by one or more file descriptors and holding state such as the current file offset and file status flags.
- · Page cache
The page cache is the kernel-managed memory that holds cached file contents for access and writeback.
- · Partitioned filter
An SST membership-filter layout split into independently loadable partitions located through a top-level index.
- · Perpendicular magnetic recording (PMR)
Perpendicular magnetic recording stores magnetization normal to the recording surface, independently of whether tracks overlap or writing is assisted.
- · Phi accrual failure detector
A phi accrual failure detector expresses suspicion as the negative base-10 logarithm of an estimated heartbeat inter-arrival tail probability.
- · PLC
Penta-level cell (PLC) NAND encodes five bits per cell using thirty-two threshold-voltage states.
- · Poisson rate model
A Poisson rate model represents an event count over a specified exposure by a Poisson distribution with mean equal to the event rate multiplied by that exposure.
- · Pole erasure
Pole erasure is unintended disturbance of recorded data by a write pole's residual magnetic field after the write current has stopped.
- · POSIX_FADV_WILLNEED
POSIX_FADV_WILLNEED is file-access advice announcing that an application expects to read a specified range soon.
- · Power-aware coding
Power-aware coding uses redundancy to serve selected reads without activating the disks containing the requested data.
- · Probability calibration
Probability calibration is agreement between forecast probabilities and conditional event frequencies among observations receiving those probabilities.
- · pwrite
pwrite is a seekable-file operation that writes up to the requested byte count at an explicit offset without changing the open file description's current offset.
- · QLC SSD
A QLC SSD uses NAND flash that encodes four bits per cell using sixteen threshold-voltage states.
- · readahead (Linux syscall)
The Linux readahead syscall requests advance loading of a file range into the page cache without advancing the open file description's offset.
- · RENAME_EXCHANGE
RENAME_EXCHANGE is a Linux renameat2 flag that atomically swaps two existing pathname bindings.
- · RENAME_NOREPLACE
RENAME_NOREPLACE is a Linux renameat2 flag that rejects a rename when its destination already exists.
- · Retention interference
Retention interference is dependence of a flash cell's charge-retention behavior on the state of a neighboring cell.
- · Ribbon filter
A static approximate-membership filter built using banded Boolean linear systems, offering a space and construction-cost tradeoff relative to Bloom filters.
- · RocksDB BackupEngine
RocksDB BackupEngine manages backup recovery points, their file metadata and restoration into a database directory.
- · RocksDB checkpoint
A RocksDB checkpoint is a captured database directory that can be opened separately from the source database.
- · RocksDB column family
A RocksDB column family is a separately configurable logical key space within one database, with its own memtables and SST files but a shared write-ahead log.
- · RocksDB read snapshot
A RocksDB read snapshot is a handle selecting a point-in-time read view within a running database.
- · Rotational latency
Rotational latency is the wait for a requested sector to pass under a hard-disk head after track positioning, with a half-revolution mean only under a uniform rotational-phase model.
- · RWF_NOWAIT
RWF_NOWAIT is a per-operation Linux I/O flag requesting that supported operations avoid waiting for unavailable data or contended resources.
- · SAS expander
A SAS expander is a device in a SAS domain that routes connections between its phys, so that one initiator port can reach many more drives than the initiator has phys; it works as a circuit switch rather than a packet router.
- · SAS PHY event counters
SAS PHY event counters are per-physical-link-endpoint counts of specified link events, including invalid DWORDs, disparity errors, synchronization loss and PHY reset problems.
- · SAS physical PHY
The SAS physical-layer endpoint containing the transceiver that connects to another physical PHY over a link.
- · SAS wide port
A SAS port containing more than one PHY, allowing separate connections across the port's links.
- · Serial ATA Tunneled Protocol
The SAS protocol that carries SATA communication through a SAS domain to an attached SATA device.
- · Serial Attached SCSI (SAS)
An enterprise storage interface that extends SCSI semantics over a serial point-to-point or expander topology.
- · Serial Management Protocol
The SAS protocol for management exchanges such as discovering and configuring expander topology.
- · Serial SCSI Protocol
The SAS protocol that carries SCSI commands, task management, data and responses between initiator and target ports.
- · Short write
A short write is a successful write operation that reports transferring a positive number of bytes smaller than the requested count.
- · SIEVE cache eviction
SIEVE is a cache replacement policy that retains insertion order and uses a visited bit and a moving pointer to select an unvisited eviction victim.
- · SQLite Online Backup API
The SQLite Online Backup API copies one named database between distinct connections using an explicit initialization, page-copy and cleanup sequence.
- · SQLite WAL end mark
A SQLite WAL end mark is the committed WAL boundary that determines which page versions a read transaction may observe.
- · Stable Bloom filter
A stable Bloom filter is an approximate stream-history filter that ages counter cells to bound memory while allowing both false positives and forgotten duplicates.
- · Superparamagnetic limit
The superparamagnetic limit is a thermal-stability constraint arising when magnetic reversal becomes too likely over the required storage time.
- · Tail latency
Tail latency describes the high-latency part of an operation response-time distribution, usually summarized using specified high percentiles.
- · Tied request
A tied request is a duplicate request whose server coordinates with the other copy's server so that starting execution can cancel or deprioritize a still-queued counterpart.
- · TLC SSD
A TLC SSD uses NAND flash that encodes three bits per cell using eight threshold-voltage states.
- · writev
writev is a vectored I/O operation that writes an ordered array of buffers to a file descriptor and reports the total number of bytes transferred.
- · Adjacent-track interference (ATI)
Adjacent-track interference is degradation of recorded HDD data caused by writes on a neighboring track.
- · Advisory file lock
An advisory file lock coordinates access among cooperating participants without itself preventing access that bypasses the locking protocol.
- · Anti-entropy repair
Anti-entropy repair compares replica state over a selected scope and synchronizes differences to reduce divergence.
- · ArmorCache
ArmorCache is an HDD cache-protection feature that uses OptiNAND flash to preserve buffered user writes during emergency power loss on supported drives.
- · Asynchronous replication
Asynchronous replication allows primary-side completion without waiting for the corresponding update to reach a specified processing stage on a replica.
- · ATI refresh
ATI refresh is a protective HDD firmware operation that reads and rewrites recorded data to mitigate write-induced adjacent-track interference.
- · Atomic rename
Atomic rename is a filesystem namespace operation whose successful replacement of an existing destination changes the file reached through that pathname without exposing an intermediate missing destination to concurrent lookups.
- · B-epsilon tree
A B-epsilon tree is an ordered index that buffers update messages in internal nodes and transfers them toward leaves in batches.
- · Barium ferrite tape
Barium ferrite (BaFe) tape is particulate magnetic tape whose coating uses hexagonal platelet barium ferrite oxide particles instead of acicular iron-cobalt metal particles, because BaFe keeps usable coercivity at much smaller particle volumes (about 1,000 to 1,600 nm3, against metal particles that lose coercivity below about 2,500 nm3).
- · Binary fuse filter
A binary fuse filter is a static approximate-membership structure that assigns XOR-linked fingerprints across consecutive segments to support compact membership tests.
- · Block storage
Block storage presents capacity as a numbered sequence of fixed-size addressable logical blocks that a host reads and writes by block address, leaving any file, directory or object semantics to software above it.
- · Bounded crash testing
Bounded crash testing systematically explores recovery behavior within an explicitly limited set of workloads, initial states and crash points.
- · Btrfs filesystem du
A Btrfs command that uses FIEMAP to report extent usage and sharing for selected files and directory trees.
- · Buffered upsert
A buffered upsert encodes a deferred transformation of a key's value without first fetching that value.
- · Byte hit ratio
Byte hit ratio is the fraction of requested payload bytes served from a cache under a stated request-accounting rule.
- · Cache admission
Cache admission is the decision to retain a fetched item in a cache or cache segment rather than bypass it.
- · Cause-specific hazard
A cause-specific hazard is the instantaneous rate of one event type among units still free of all modeled first events.
- · CFexpress
CFexpress is a removable memory-card standard of the CompactFlash Association that carries the NVMe protocol over PCIe, in three form factors (Type A, B and C with 1, 2 and 4 PCIe lanes), aimed at professional imaging and embedded recorders.
- · Checksum
A checksum is a value computed from specified data and later compared with a reference value to detect changes.
- · CKD DASD
Count-key-data (CKD) is the track format of classic mainframe direct access storage devices: each track holds records made of a count area (record identifier and lengths), an optional key area and a data area, separated by gaps, instead of the fixed-size sectors of SCSI-style disks.
- · Cold archive tier
A cold archive tier is the storage tier for data that is rarely read, for example once a year or less, and kept for long retention, so its hardware is provisioned for low power and capacity rather than low-latency access, accepting first-byte delays of seconds or more.
- · Command duration limit
A command duration limit is a device-side policy selected by a descriptor index for supported storage commands.
- · Competing risks
Competing risks are mutually exclusive first-event types for which occurrence of one prevents the others from being the first event in the defined process.
- · Composite checksum
A composite checksum is an object-level integrity value formed by combining the ordered checksums of the object's upload parts under a specified construction.
- · Computational storage drive
Computational storage is an architecture that couples computation to storage to offload host processing or reduce data movement; a computational storage drive (CSD) is a storage element that contains one or more computational storage engines (such as a CPU or FPGA) together with persistent data storage, and can still work as a standard drive.
- · Concurrent positioning range (ATA)
An ATA concurrent positioning range is a contiguous logical-block range reported as supporting positioning concurrently with other reported ranges.
- · Content-defined chunking
Content-defined chunking divides a byte stream using content-dependent boundary rules rather than only fixed offsets.
- · Conventional magnetic recording (CMR)
Conventional magnetic recording (CMR) uses non-shingled tracks that permit in-place sector updates without SMR-style band rewriting.
- · Copy-on-abundant-write
Copy-on-abundant-write is a research cloning policy that buffers changes to postpone unsharing until enough work accumulates to justify copying.
- · Crash-test oracle
A crash-test oracle is a checker that decides whether a recovered state satisfies the specified correctness and durability contract.
- · Cumulative incidence function
A cumulative incidence function gives the probability that a specified event type occurs by a given time before any competing event.
- · Curie-temperature distribution
Curie-temperature distribution is the spread of ferromagnetic transition temperatures across the grains of a recording medium.
- · DAT cartridge
A DAT cartridge is a 3.81 mm wide magnetic tape cassette recorded by helical scan in the DDS (Digital Data Storage) format, standardised by ECMA from 1990 and later adopted by ISO/IEC, which grew from 60 m and 90 m tapes to 125 m tapes holding 12 GB uncompressed in the DDS-3 generation.
- · Degaussing
Degaussing uses a strong reverse magnetic field matched to the medium's coercivity to disrupt recorded magnetization, and can sanitize suitable magnetic media such as tape and hard disk platters but not flash.
- · Delayed allocation
A filesystem allocation strategy that defers choosing physical storage locations for data already accepted into its buffered write path.
- · Disk shelf
A disk shelf is a rack-mounted enclosure that provides power, cooling and SAS connectivity to a group of drives and reports its state through SCSI Enclosure Services; one open hardware specification, for example, describes a 2U unit holding 30 3.5-inch drives.
- · DLT
Digital Linear Tape (DLT) is a family of half-inch (12,7 mm) single-reel linear serpentine tape cartridges and drives, standardised by ECMA as DLT 1 to DLT 6 and extended by later Super DLT and DLT-S generations, used mainly for mid-range backup before LTO became the more common choice.
- · Doublewrite buffer
An InnoDB doublewrite buffer stages data-page copies before their final file writes so crash recovery can repair an incomplete page write.
- · Drive firmware revision
A drive firmware revision is the vendor-assigned version identifier reported for the firmware running on a storage drive.
- · Dual-actuator HDD
A dual-actuator HDD is a hard drive with two independently positioning head assemblies that can service different data regions concurrently.
- · Dual-port drive
A dual-port drive exposes two independent ports (two SAS ports with separate SAS addresses, or an NVMe subsystem with more than one port and two or more controllers) to the same media, so a host can keep reaching the data over a second path in multipath arrays.
- · Dual-stage actuator (HDD)
A dual-stage HDD actuator combines a coarse positioning stage with a fine positioning stage to control the read/write head.
- · DWPD
DWPD is an SSD endurance rating that normalizes cumulative host writes by host-addressable drive capacity and the number of days in a specified period.
- · End-to-end data protection
End-to-end data protection is an integrity scheme whose generation and verification points span an explicitly defined data path.
- · Energy-assisted perpendicular magnetic recording (ePMR)
Energy-assisted perpendicular magnetic recording (ePMR) is Western Digital's write-assist approach in which a DC bias through the write pole makes the write field more consistent, reducing recording jitter.
- · Error recovery control
Error recovery control configures read and write recovery time limits for applicable ATA commands.
- · ESCON
ESCON (Enterprise Systems Connection) is IBM's fiber-optic, half-duplex serial channel architecture for mainframes, built to replace the copper bus-and-tag parallel channel and later superseded by FICON.
- · ESDI
ESDI (Enhanced Small Device Interface, standardised as X3.170-90) is a disk-drive interface in which one controller drives up to seven devices over a daisy-chained control cable that carries serial commands and status, plus a separate radial data cable per drive carrying NRZ data and clocks; the standard text covers 0 to 24 Mbit/s and both magnetic and optical disks.
- · Failure burst
Under an adjacent-gap rule, a failure burst is a maximal sequence of event starts in which every consecutive gap is no greater than a specified window.
- · Failure domain
A failure domain is a group of components that can be affected by the same failure event because they share a dependency or physical location.
- · Failure label
A failure label is a recorded outcome assigned to an observation under an explicit rule defining which event counts as failure and when it occurs.
- · Failure operation ticket
A failure operation ticket is a management-system record of a detected hardware problem and its handling status, rather than necessarily a unique failed component.
- · False positive rate
The false positive rate is the fraction of observed negative cases incorrectly classified as positive at a specified decision threshold.
- · FastCDC
FastCDC is a content-defined chunking approach that combines Gear-based boundary selection, sub-minimum skipping and chunk-size normalization.
- · Fault model
A fault model specifies the failures a test or analysis admits, their scope and timing, and which system state and components remain available afterward.
- · Fibre Channel
Fibre Channel (FC) is a serial networking technology standardised by INCITS committee T11, built on frames exchanged between ports over point-to-point links, switched fabrics or loops, and widely used to carry SCSI (and NVMe) traffic between servers and storage in a SAN.
- · FICON
FICON (Fibre Connection) is the IBM mainframe I/O channel protocol that runs as a Fibre Channel upper-level protocol (the FC-SB family), carrying channel commands and data between mainframe channels and storage control units such as DASD and tape.
- · FIEMAP
A Linux ioctl interface that reports file extent mappings and associated state flags over a requested logical byte range.
- · File extent
A filesystem mapping for a contiguous range of file blocks and the storage blocks backing that range.
- · File preallocation
Reserving filesystem allocation for a file range before the application writes that range's data.
- · File storage
File storage exposes data as named files in a hierarchical directory tree that many clients access over a network file-sharing protocol such as NFS (registered port 2049) or SMB, so clients operate on files rather than on raw blocks.
- · Filesystem defragmentation
Reorganizing a file's storage extents to reduce fragmentation while preserving its logical contents.
- · Fixed block architecture
Fixed block architecture (FBA) is a mainframe disk format in which every data record has the same length (512 bytes on the devices documented here) and each block is addressed by a sequential block number, instead of by cylinder, head and record number as in count-key-data (CKD) devices.
- · Flash data retention
Flash data retention is the ability of programmed NAND data to remain recoverable over time without refresh under specified wear, storage and read conditions.
- · Flash module
A flash module is a replaceable unit that packages NAND dies and a controller on one board or in one package, used in arrays and servers either in a standard form factor (for example E1.S, 111.49 mm long) or in a custom card shape.
- · Fluid dynamic bearing
A fluid dynamic bearing (FDB) in an HDD spindle motor carries the rotating shaft or hub on a thin pressurised film of lubricant (usually oil, sometimes air) generated by grooves such as herringbone patterns, so that the rotor and stator do not touch while spinning, replacing the rolling balls of a ball-bearing spindle.
- · Flux-control microwave-assisted magnetic recording (FC-MAMR)
Flux-control microwave-assisted magnetic recording (FC-MAMR) is Toshiba's HDD write-assist technique that uses a spintronic device between the write poles to redirect magnetic flux toward the recording medium and strengthen the write field.
- · Flying height
Flying height is the physical clearance between a flying HDD head/slider region and the disk surface.
- · Full-object checksum
A full-object checksum is an integrity value computed for an object's complete ordered byte sequence rather than for a list of its part checksums.
- · Glass platter substrate
A glass platter substrate is a disk blank made of chemically strengthened glass or glass-ceramic instead of nickel-phosphorus-plated aluminium alloy; it is stiff, hard, polishes to a very smooth surface and expands far less with temperature, but it is brittle, a poor heat conductor and electrically insulating, which brings its own processing problems.
- · GMR read head
A GMR read head is a hard-disk read element that uses giant magnetoresistance in a stack of thin magnetic and non-magnetic layers to turn very weak magnetic changes from the disk into large changes in electrical resistance; the first GMR read head was launched in 1997 and enabled the areal density gains of the late 1990s.
- · HAMR thermal gradient
A HAMR thermal gradient is the spatial rate of temperature change in the recording medium along a specified direction.
- · Hazard rate
The hazard rate is the instantaneous event rate at a given time conditional on the event not having occurred before that time.
- · HDD head gas bearing
An HDD head gas bearing is the thin moving gas film whose pressure supports the head-carrying slider against its suspension load.
- · Head-disk interface (HDI)
The head-disk interface is the interacting region between an HDD's head-carrying slider and rotating disk surface.
- · Helical scan recording
Helical scan recording writes short, slanted tracks across a tape that moves slowly past a rotating head drum (scanner), so the heads move far faster relative to the tape than the tape moves; it is used in DAT-family data cartridges and in some video formats.
- · Helium leak port
Helium leak port is an unverified legacy corpus label, not an established name for a lifetime diagnostic interface on helium HDDs.
- · Helium-filled HDD
A helium-filled HDD is a hard disk drive whose head-disk assembly operates in a sealed helium atmosphere to reduce aerodynamic loading.
- · Hint bit
A PostgreSQL hint bit is a tuple-header flag caching transaction-status information used to determine row visibility.
- · Host bus adapter (HBA)
A host bus adapter (HBA) is the hardware bridge between a server's internal I/O bus (for example PCI Express) and a storage transport such as SCSI, SAS, SATA or Fibre Channel: it contains the initiator ports that send commands to the storage devices (targets).
- · Host-managed band (SMR)
In host-managed shingled magnetic recording, the host (not the drive) tracks which sequentially written bands are full or empty and must write each one strictly at its write pointer; the standards call these bands zones, and the strict kind is the sequential write required zone of a host-managed zoned block device.
- · Host-side flash GC
Host-side flash garbage collection is the reclamation of free space by host software, such as a log-structured file system, a host-based FTL or an application, which copies still-valid data out of a victim region and then explicitly resets or erases that region instead of leaving the work to the SSD's internal FTL.
- · Hot standby feedback
Hot standby feedback is PostgreSQL's mechanism for reporting standby query retention needs upstream to defer conflicting row cleanup.
- · Hot storage tier
A hot storage tier holds data that is requested frequently and must be served with low latency, on media and with a layout chosen for I/O operations per second rather than storage efficiency; in a published large-scale blob-storage study, content under one day old received more than 100 times the request rate of one-year-old content.
- · Hybrid flash array
A hybrid flash array is a storage system that combines flash devices with hard-disk capacity pools and uses caching or tiering software to keep frequently accessed data on flash while most of the data stays on disks.
- · I/O fencing
I/O fencing enforces exclusion of a node from shared storage so that recovery cannot be undermined by its continued access.
- · I/O timeout
An I/O timeout is expiration of a layer's waiting limit for an operation, which can trigger recovery without proving that the operation never executed.
- · IBM RAMAC
RAMAC (Random Access Method of Accounting and Control) was the first commercial moving-head disk storage product: a stack of fifty 24-inch disks with air-bearing heads, delivered in 1956 as the storage unit of the 305 RAMAC accounting system, giving random access to about 5 million characters (3.75 MB equivalent) where earlier systems used cards, tape or drums.
- · Idle Read After Write (IRAW)
Idle Read After Write is a research technique that retains completed-write data for comparison with disk contents during idle periods.
- · Independent access range (Linux)
A Linux independent access range is an advertised region of a block device that can service requests in parallel with its other reported regions.
- · Informative censoring
Informative censoring occurs when the observation-ending process conveys information about the event time beyond what the survival analysis accounts for.
- · Interval-censoring
Interval-censoring is incomplete event-time observation in which an event is known to have occurred within an interval but its exact time is unknown.
- · Invertible Bloom lookup table
An invertible Bloom lookup table is a randomized data structure that supports updates and can list encoded entries with high probability under its sizing assumptions.
- · JBOD
JBOD (Just a Bunch Of Disks) originally meant disks used without coordinated control software and now usually means an enclosure of drives, with or without RAID functions; for software-managed storage the drives are typically presented individually through a host bus adapter, leaving redundancy to host software.
- · JEDEC SSD standards
The JEDEC SSD standards are JESD218 (SSD requirements and endurance test method) and JESD219 (SSD endurance workloads), developed by the JC-64.8 subcommittee, which define the TBW endurance rating and how it is verified for client and enterprise SSD classes.
- · Kaplan-Meier estimator
The Kaplan-Meier estimator is a nonparametric estimator of the survival function that multiplies conditional survival estimates at observed event times while accounting for right-censored observations.
- · Key-value SSD
A key-value SSD is a solid-state drive that stores and retrieves variable-length values addressed by a key (up to 16 bytes in the NVMe Key Value command set) through Store, Retrieve, Delete, Exist and List commands, instead of reading and writing fixed-size logical blocks.
- · LMDB
LMDB is an embedded transactional key-value database that uses memory-mapped B+ trees, copy-on-write updates and multi-version concurrency control.
- · LMDB reader table
The LMDB reader table records reader process, thread and transaction identities so page reclamation can respect active snapshots.
- · Log sequence number
A log sequence number identifies a position in an engine's ordered recovery-log address space.
- · Log-structured merge-tree
A write-oriented index that buffers updates and merges sorted components over time, trading write batching against read and compaction work.
- · Logical decoding
Logical decoding reconstructs database changes from a storage-level log into a representation that a consumer can interpret without the database's physical storage format.
- · Logical failover slot
A PostgreSQL logical failover slot is a logical replication slot enabled for synchronization to a physical standby so consumption can resume after promotion.
- · Logical unit (LUN)
A logical unit (LU) is an addressable I/O device inside a SCSI target, such as a volume presented by a storage array, and its address within that target is the logical unit number (LUN), a 64-bit field in the iSCSI header.
- · LTO (Linear Tape-Open)
An open tape format standard for high-capacity cartridges, advanced by the LTO Program for backup and archive.
- · M.2
M.2 is a small module and connector form factor, originally called Next Generation Form Factor, that carries SATA or PCI Express (including NVMe) SSDs as well as wireless cards; the key notches and the length code of a module, not its name, tell which sockets it fits and which interface it may use.
- · Magnetic tape storage
Sequential storage on linear magnetic tape in removable cartridges, used mainly for backup and cold archive.
- · Mainframe DASD
Mainframe DASD (direct-access storage device) is the disk storage attached to mainframe systems through channel programs, historically organised as count-key-data tracks addressed by cylinder, head and record, with fixed-block volumes as a second format.
- · Memristor storage
A memristor is a circuit element proposed in 1971 that relates charge and magnetic flux; the term is also used for two-terminal thin-film devices whose resistance depends on past current and is changed by moving ions, which are researched as dense non-volatile memory (resistive RAM). The sources used here treat it as an emerging, researched memory technology, and it is not mainstream bulk storage.
- · Merkle tree
A Merkle tree combines hashes of data and child nodes into a root so that selected entries or subtrees can be checked without transferring the entire data set.
- · Metal particle tape
Metal particle (MP) tape is magnetic tape whose recording layer is made of metal particles in a binder; the first LTO format (Ultrium-1, ECMA-319) requires such a metal-particle coating on 12.65 mm tape, before barium ferrite came into use from LTO-6.
- · Microwave-assisted magnetic recording (MAMR)
Microwave-assisted magnetic recording (MAMR) denotes HDD write-assist approaches whose implementation must be distinguished: MAS-MAMR uses a microwave magnetic field to assist switching in the medium, while Toshiba's FC-MAMR strengthens the write field through flux control.
- · Microwave-assisted switching MAMR (MAS-MAMR)
Microwave-assisted switching MAMR (MAS-MAMR) is a recording technique in which a spin-torque oscillator adds a microwave magnetic field to the write field, reducing the field needed to reverse magnetization in the recording medium.
- · Mission-critical HDD
A mission-critical HDD is a high-rotation-speed (10,000 or 15,000 rpm) enterprise hard drive, usually with a SAS interface and a small capacity per drive, built for low rotational latency and data-integrity features in transactional workloads that were served by spinning disks before flash.
- · Multi-level cell (MLC)
NAND flash that stores two bits per cell using four distinguishable threshold-voltage states.
- · Multipart upload
Multipart upload is an object-transfer process that sends numbered parts independently and assembles them into one object when the upload is completed.
- · NAND die stacking
NAND die stacking means putting several NAND flash dies into one package so that they share the package's pins and I/O bus while each die is selected by its own chip-enable signal, which raises the capacity per package.
- · NAND dwell time
NAND dwell time is the interval between specified successive program or erase stress events on the same cells, during which partial self-recovery can occur.
- · NAND self-recovery
NAND self-recovery is the partial relaxation of cycling-induced cell degradation during time between stress events.
- · NAND thermal annealing
NAND thermal annealing is a controlled heat treatment studied for partially reversing cycling-induced degradation in flash-memory cells.
- · NAS
Network-attached storage (NAS) is a storage device or server that connects to an IP network and gives clients file-level access to a file system it manages, using file-sharing protocols such as NFS or SMB, rather than attaching to a host as block devices.
- · Near-field transducer (NFT)
A near-field transducer is a nanoscale optical structure that couples supplied light into a confined near field, used in HAMR to heat the recording medium.
- · Nearline SAS
Nearline SAS (NL-SAS) is a vendor class of high-capacity, typically 7,200-rpm 3.5-inch hard drives that use the SAS interface and SCSI command set, intended for always-on bulk capacity tiers rather than latency-sensitive primary storage.
- · NFS open-file preservation
NFS open-file preservation is the NFSv4.1 OPEN-result promise to retain a file removed while open, including through the server-restart grace period for open reclamation.
- · NFS silly rename
NFS silly rename is a client workaround that temporarily renames an in-use file's directory entry and defers its removal until local references are released.
- · NVDIMM-N
NVDIMM-N is a standard DIMM that works as ordinary DRAM during normal operation and, when power is lost, copies its contents to flash on the same module using stored energy (a supercapacitor or similar), then restores them at the next boot; the host sees DRAM-speed, byte-addressable memory whose size is limited to the DRAM on the module.
- · NVM subsystem
An NVMe system containing one or more controllers, domains and ports, with zero or more namespaces.
- · NVMe backplane
An NVMe backplane is the board in a server or storage enclosure that carries PCI Express lanes, power and sideband signals from the host (or from a PCIe switch) to a row of drive-bay connectors, for example U.2 (SFF-8639) receptacles or EDSFF card-edge slots.
- · NVMe base specification
The NVM Express Base Specification is the transport-independent core document of the NVMe family: it defines controller properties, the admin command set, queue mechanics and the common aspects of I/O command sets, while transport bindings and individual I/O command sets are separate documents.
- · NVMe Composite Temperature
NVMe Composite Temperature is an implementation-specific temperature value in kelvins representing the controller and its associated namespaces in the SMART / Health Information log.
- · NVMe Critical Composite Temperature Time
NVMe Critical Composite Temperature Time is a cumulative SMART / Health log duration in minutes of controller operation with Composite Temperature at or above the critical composite-temperature threshold.
- · NVMe Critical Warning
NVMe Critical Warning is the SMART / Health Information log bitmask identifying controller warning conditions at the time the log is read.
- · NVMe Host Controlled Thermal Management
NVMe Host Controlled Thermal Management is a controller feature through which a host configures temperature targets for automatic transitions to lower-power active power states or vendor-specific thermal actions.
- · NVMe Host Identifier
The NVMe Host Identifier identifies which controllers in an NVM subsystem belong to the same host for purposes including reservation rights.
- · NVMe multipath
Using multiple routes to the same NVMe namespace for path failover and, where supported, load distribution.
- · NVMe namespace
An NVMe resource collection that a host can address through a namespace identifier, such as formatted non-volatile storage.
- · NVMe namespace ID (NSID)
An NVMe namespace ID (NSID) is a 32-bit handle used by a controller and host to address a namespace, not a globally unique namespace identity.
- · NVMe namespace identifiers
NVMe namespace identifiers are the NGUID, EUI64 and namespace UUID values used to distinguish a namespace independently of its command-addressing NSID.
- · NVMe reservation
An NVMe reservation applies a selected host-access policy to an NVMe namespace.
- · NVMe reservation persistence
NVMe reservation persistence determines whether a namespace retains its reservations and registrations across power loss.
- · NVMe SMART log
The NVMe SMART log is the standardized log page containing controller health indicators and cumulative usage, endurance and error counters.
- · NVMe subsystem NQN
An NVMe subsystem NQN is the NVMe Qualified Name identifying an NVM subsystem, distinct from a host NQN and from individual namespace identifiers.
- · NVMe Warning Composite Temperature Time
NVMe Warning Composite Temperature Time is a cumulative health-log duration in minutes governed by composite-temperature thresholds and, in supporting revisions and configurations, temperature hysteresis.
- · NVMe-MI
NVMe-MI is the NVMe Management Interface specification: it defines a command set and message format for monitoring and managing NVMe SSDs and enclosures out of band over MCTP (on SMBus/I2C, I3C or PCIe vendor-defined messages) and in band through NVMe Admin commands, without using the host data path.
- · Open NAND block
An open NAND block is a partially programmed erase block whose remaining pages have not yet completed programming.
- · Open-file-description lock
An open-file-description lock is a byte-range file lock owned by an open file description rather than by a process.
- · Optical jukebox
An optical jukebox is a robotic library in which a picker moves removable optical discs (CD, DVD, magneto-optical or Blu-ray) between storage slots and one or more drives, so that many more discs than drives can be reached without an operator.
- · OptiNAND
OptiNAND is a flash-assisted HDD architecture that integrates an embedded iNAND device with disk-controller hardware and firmware.
- · PATA
Parallel ATA (PATA) is the original AT Attachment interface, in which a host reaches up to two disk drives per cable through a 40-pin connector that carries a compatible register set and a 16-bit parallel data bus, before serial ATA replaced it.
- · PCIe 4.0 storage
PCIe 4.0 is the fourth generation of the PCI Express interconnect, signalling at 16 GT/s per lane with 128b/130b encoding (twice the rate of PCIe 3.0), which sets the link bandwidth available to PCIe-attached storage such as NVMe SSDs.
- · PCIe 5.0 storage
PCIe 5.0 is the fifth generation of the PCI Express interface, running each lane at 32.0 GT/s with 128b/130b encoding, twice the per-lane rate of PCIe 4.0, which gives NVMe storage devices up to about 63 GB/s per direction on a x16 link and about 15.75 GB/s on the x4 link typical of an SSD (our calculation).
- · PCIe retimer
A PCIe retimer is a physical-layer, protocol-aware and software-transparent extension device that fully recovers the data and clock of an incoming PCIe signal and retransmits a fresh copy, so one long channel becomes two separate, independently compliant electrical link segments; it is used where traces, connectors or cables would otherwise exceed the PCIe loss budget, for example in servers with many NVMe slots.
- · Perpendicular magnetic tape
Perpendicular magnetic tape is magnetic tape whose recording layer is magnetised perpendicular to the tape surface (as in perpendicular recording on disks), a medium that laboratory work has used to demonstrate areal densities of 148 to 201 Gb/in2.
- · Phase-change memory (storage)
Phase-change memory (PCM) stores a bit as the resistance of a small volume of chalcogenide alloy: a short, high current pulse melts and quenches it into a high-resistance amorphous state, a longer, moderate pulse crystallises it into a low-resistance state, and a small current reads the resistance without changing it.
- · posix_fallocate
A POSIX library function that reserves storage for a file byte range and reports failure by returning an error number directly.
- · Power loss protection
Power-loss protection is a set of hardware and firmware measures intended to preserve specified storage-device data and internal state when external power is unexpectedly removed.
- · Power-on hours
Power-on hours is accumulated powered-operation time reported for a device, interpreted according to that device's counter and interface semantics.
- · Precision
Precision is the fraction of positive predictions that match observed positive outcomes at a specified decision threshold.
- · Prediction cadence
Prediction cadence is the schedule or interval at which a prediction system produces successive predictions for the same monitored entity.
- · Prediction horizon
A prediction horizon is the future interval after a prediction time in which a specified target event is evaluated.
- · Process-associated record lock
A process-associated record lock is a byte-range file lock whose ownership is attached to a process rather than to an open file description.
- · Program/erase cycle
One complete program and erase of a NAND erase block, a core endurance metric.
- · Pseudo-SLC (pSLC)
Operation of multi-bit-capable NAND in one-bit-per-cell mode, trading storage density for single-bit behavior.
- · Quad-level cell (QLC)
NAND flash that stores four bits per cell using sixteen threshold-voltage states.
- · Racetrack memory
Racetrack memory is a proposed non-volatile memory in which bits are stored as a train of magnetic domains, separated by domain walls, along a magnetic nanowire and are shifted past a fixed write element and read element by current pulses; it is a research concept, not a shipping bulk store.
- · RAIDZ
RAIDZ is ZFS's distributed-parity virtual-device layout with one, two or three parity fragments and block-dependent stripe widths.
- · RAIDZ expansion
RAIDZ expansion widens an existing RAIDZ virtual device by attaching another member and redistributing allocated space while retaining old blocks' data-to-parity ratios.
- · Rate of occurrence of failures
The rate of occurrence of failures (ROCOF) is the derivative with respect to time of the expected cumulative failure count of a repairable system.
- · Rateless IBLT
Rateless IBLT is a set-reconciliation scheme that streams additional coded symbols until the receiver can recover the set difference without first estimating its size.
- · RDMA storage networking
RDMA storage networking moves storage commands and data between hosts and storage systems with remote direct memory access, so data is placed directly into registered memory buffers without intermediate data copies; NVMe over Fabrics (NVMe/RDMA) and iSCSI Extensions for RDMA (iSER) are the standardised examples.
- · Read-after-write verification
Read-after-write verification checks stored data after a write, with comparison guarantees determined by the verification mechanism.
- · Recall
Recall is the fraction of observed positive outcomes identified by a classifier at a specified decision threshold.
- · Reflink
Filesystem support for sharing physical extents between files until one copy is modified.
- · Replication lag
Replication lag is a difference in replication progress between a source and a replica, measured at a named processing stage in a stated unit.
- · Replication slot
In PostgreSQL, a replication slot tracks a consumer's progress and retains WAL and other resources needed for that consumer to continue.
- · Right-censoring
Right-censoring records that a time-to-event exceeds the last observed event-free time without observing the event time itself.
- · Robotic tape loader
A robotic tape loader is the mechanism in a tape library (the picker, or medium transport element in SCSI terms) that moves cartridges between storage slots, import/export slots and tape drives under host command.
- · Rotational media
Rotational media is storage that keeps data on one or more spinning platters or discs, read and written by actuator-positioned heads, such as a hard disk drive or an optical disc, so access time depends on moving parts and on the angular position of the data.
- · Run-length limited (RLL) encoding
Run-length-limited (RLL) encoding is a modulation code for magnetic and optical recording that limits the run of 0s between successive 1s (flux transitions) to at least d and at most k, so that transitions are not too close for peak detection and not too far apart for clock recovery; in disk drives it was later replaced by codes designed for PRML detection.
- · S.M.A.R.T.
Self-Monitoring, Analysis and Reporting Technology exposes drive health metrics to the host.
- · SAN
A storage area network (SAN) is a storage network that gives multiple servers (initiators) block-level access to shared storage devices (targets) over a fabric such as Fibre Channel, iSCSI on TCP/IP or NVMe over Fabrics.
- · SCSI persistent reservations
SCSI persistent reservations control initiator access to a logical unit using registered keys and a selected reservation type.
- · SCSI VERIFY
SCSI VERIFY checks selected logical blocks using verification semantics that depend on command fields and device behavior.
- · Seek time
Seek time is the time a hard disk's actuator needs to move the head stack to the target track, made up of acceleration, coasting, deceleration and a final settling phase; it excludes the rotational delay and transfer time that follow.
- · Set reconciliation
Set reconciliation is the task of identifying items present in one of two sets but not the other.
- · Single-level cell (SLC)
NAND storage of one bit per cell using two distinguishable threshold-voltage states.
- · SLC cache
A NAND region operated in one-bit mode as a write cache for denser NAND storage, with still-valid data eligible for later migration.
- · SLC SSD
An SLC SSD is a solid-state drive whose NAND flash stores one bit per cell, using only two threshold-voltage states separated by one wide guard band, which gives the highest endurance and the lowest program latency of the cell types at the cost of the least capacity per cell.
- · SNIA
SNIA (Storage Networking Industry Association) is a not-for-profit industry association that publishes storage and data-infrastructure specifications, technical positions and educational material, including the Emerald power-efficiency measurement specification that a government energy-labelling programme uses as its test method.
- · Solid-state media
Solid-state media is persistent storage built from semiconductor memory cells, today mainly NAND flash and, experimentally, emerging memories such as phase-change memory, rather than from spinning disks or moving tape.
- · Sparse file
A file with unallocated ranges that read as zeros, allowing its logical size to exceed its allocated storage.
- · SQLITE_BUSY_SNAPSHOT
SQLITE_BUSY_SNAPSHOT is SQLite's extended result code for a WAL read transaction that cannot become a writer because its snapshot is obsolete.
- · SSD controller
An SSD controller is the system-on-chip inside a solid-state drive that connects the host interface to the NAND flash through several parallel flash channels and runs the error correction and flash translation firmware that make imperfect NAND usable as a block device.
- · SSD thermal throttling
SSD thermal throttling is a controller response that limits activity or performance to manage device temperature.
- · Storage virtualization
Storage virtualization abstracts or hides the internal functions of storage devices or services from hosts and applications, for example by aggregating several physical disks or arrays into pooled logical volumes (block aggregation) that hosts see as one device, in the host, in the network or in the storage system.
- · Storage-class memory (SCM)
Storage-class memory (SCM) is non-volatile solid-state memory that sits between DRAM and block storage in latency, is often byte-addressable, and is used as a persistence or capacity tier close to the CPU.
- · Striping (storage)
Striping splits a logical block address range into fixed-size chunks and assigns consecutive chunks to different member devices in rotation, so that large or concurrent I/O can be served by several devices in parallel.
- · syncfs
syncfs is a Linux system call that synchronizes pending file data and metadata for the filesystem selected by an open file descriptor.
- · Tape buffer synchronization
Tape buffer synchronization is completion of a supported operation that confirms previously buffered tape data and marks have been written to the medium.
- · Tape encryption
Tape encryption is the encryption of data blocks by the tape drive (or a host or appliance) before they are recorded on a cartridge, in LTO drives typically with 256-bit AES in Galois/Counter Mode, so that a cartridge read without the key is unreadable.
- · Tape filemark
A tape filemark is a recorded logical delimiter separating files in a sequential tape stream.
- · Tape library
A tape library is an automated system in which a robotic mechanism moves tape cartridges between storage slots and one or more tape drives, so that many cartridges can be mounted without an operator.
- · TBW
TBW (terabytes written) is an SSD endurance rating: the total amount of host data a drive is rated to accept over its life. It follows from the flash program/erase limit, the capacity and the write amplification, and field data show the rating is conservative, not a point of failure.
- · TCG Opal
TCG Opal is a Trusted Computing Group specification for managing hardware encryption on self-encrypting storage devices: it defines authorities and credentials, locking ranges that can be locked and erased independently, and a pre-boot shadow area, and is carried over ATA or NVMe security commands.
- · Thermal writability
Thermal writability is the requirement that a magnetic write field bias the intended state strongly enough relative to thermal fluctuations to meet a recording-error target.
- · Thunderbolt storage
Thunderbolt storage is external storage, typically an NVMe enclosure, that attaches over a Thunderbolt or USB4 link by tunneling PCIe, so the drive appears to the host as a PCIe device and not as a USB mass-storage device, with access control and DMA protection needed because the device can act as a bus master.
- · TinyLFU
TinyLFU is a cache admission policy that compares approximate recent access frequencies of a candidate and an eviction victim.
- · TMR read head
A TMR read head senses recorded bits with a magnetic tunnel junction: a free layer and a pinned layer separated by a thin insulating barrier (for example MgO) through which the current tunnels, so the junction resistance changes with the relative magnetization of the two layers.
- · Torn write
A torn write is an interrupted write that persists only part of its intended update, potentially leaving a database page with mixed old and new contents.
- · Transition jitter
Transition jitter is variation in the positions of recorded magnetization reversals relative to their intended or estimated locations.
- · Transparent filesystem compression
Filesystem-managed compression of eligible file data that preserves the original logical bytes exposed to applications.
- · Triple-level cell (TLC)
NAND flash that stores three bits per cell using eight distinguishable threshold-voltage states.
- · UDO
Ultra Density Optical (UDO) is a standardised 130 mm two-sided optical disk cartridge using a 405 nm laser and phase-change recording, with rewritable (RW) and write-once (WORM) types; the first generation holds a nominal 30 Gbytes per cartridge (ECMA-350) and the second generation 60 Gbytes (ECMA-380).
- · Unlink
Unlink is the removal of a directory entry naming a file; POSIX semantics preserve existing open references, while network-filesystem behavior requires additional qualification.
- · Unwritten extent
An allocated filesystem extent whose data is not initialized and is exposed as zeros when read through the filesystem.
- · V-NAND
V-NAND (vertical NAND) is the trade name one manufacturer uses for its 3D NAND flash, in which memory cells are stacked in dozens of word-line layers on one die instead of being laid out in a single plane.
- · WAL compression
PostgreSQL WAL compression optionally compresses full-page images within its write-ahead log.
- · Warm storage tier
A warm storage tier holds online data whose request rate has fallen well below that of new data but is not zero, on capacity-optimised storage (for example erasure-coded disks sized for about 20 IOPS per terabyte), between the hot tier and cold or archive storage that can take hours or days to retrieve.
- · Wear leveling
Wear leveling is flash management that distributes program/erase activity among blocks to reduce uneven wear.
- · Workload rate limit
A workload rate limit is a manufacturer-specified HDD operating workload bound, commonly expressed as host reads plus writes in terabytes per 8,760 power-on hours.
- · Write combining
Write combining here means collecting small or scattered writes in a buffer and issuing them as fewer, larger, ordered operations, so that flash sees fewer partial-block merges and shingled disks see fewer read-modify-write rewrites of a band; it reduces write amplification but needs a buffer that survives power loss.
- · Write-ahead log
A write-ahead log is a recovery log whose relevant records must become durable before the corresponding changes are persisted in database data files.
- · Write-back cache
Write-back caching completes a write after updating the cache and defers updating its backing store until later.
- · XFS Active Item List
The XFS Active Item List tracks logged items whose journal records remain needed until later retirement.
- · XFS deferred operation
An XFS deferred operation postpones metadata work to a later transaction to satisfy locking or reservation constraints.
- · XFS delayed logging
XFS delayed logging aggregates metadata changes in memory before committing them to the filesystem journal.
- · XFS intent log item
An XFS intent log item describes deferred work that log recovery may need to complete.
- · XFS relogging
XFS relogging records an object's outstanding metadata changes again in a newer transaction.
- · XFS rolling transaction
An XFS rolling transaction commits one step of a transaction chain and continues with a fresh transaction using a permanent log reservation.
- · XOR filter
An XOR filter is a static approximate-membership structure whose lookup compares a key fingerprint with the XOR of shared array entries.
- · ZAC
ZAC (Zoned Device ATA Command Set) is the ATA command set through which a host identifies and controls the zones of host-aware or host-managed SMR disks on a SATA interface, the counterpart of the SCSI ZBC command set.
- · ZBC
ZBC (Zoned Block Commands) is the T10 SCSI command set for zoned block devices, whose host-managed model reports peripheral device type 14h and requires writes in sequential-write-required zones to start at the zone's write pointer.
- · ZFS error scrub
A ZFS error scrub is an integrity scan restricted to files already recorded with data errors.
- · Zone Append
The native NVMe Zone Append command writes at a controller-selected position within a specified zone and returns the assigned starting logical block address on successful completion.
- · Zone write locking
A Linux mechanism that serializes covered write requests to the same sequential zone using a per-zone lock.
- · Zone write plugging
A Linux block-layer mechanism that queues covered writes per sequential zone before dispatch to preserve the issuer's sequential write pattern.
- · Zoned block device (ZBD)
A zoned block device exposes a logical block address space divided into zones with zone-specific access rules.
- · Adaptive Replacement Cache
Adaptive Replacement Cache adjusts a recency/frequency balance using resident lists and nonresident histories.
- · Adaptive scrubbing
Adaptive scrubbing adjusts the rate or schedule of proactive storage verification in response to observed conditions or estimated error risk.
- · AHCI
Advanced Host Controller Interface, the PCI class driver model SATA SSDs and HDDs used before NVMe dominance.
- · ANA group
An ANA group is a set of NVMe namespaces whose ANA access states change together for a given controller.
- · Application quiescing
Application quiescing is application-specific preparation and temporary restriction of writes to establish a defined state for snapshot or backup capture.
- · Application-consistent snapshot
An application-consistent snapshot captures an application's data through a coordinated procedure that satisfies that application's consistency requirements.
- · Asymmetric Namespace Access
Asymmetric Namespace Access is the NVMe mechanism for reporting namespace access states through different controllers.
- · Asynchronous commit
Asynchronous commit is an acknowledgment mode that can report a transaction as successful before its recovery log has reached durable storage.
- · ATA SMART
ATA SMART is the ATA interface for retrieving drive self-monitoring attributes, status and logs from supporting devices.
- · Authenticated encryption
Authenticated encryption combines confidentiality with cryptographic integrity protection for the protected message.
- · Background fsck
Background fsck is a filesystem-specific consistency-checking mode that performs permitted repairs while the filesystem remains in use.
- · BD-R HTL
BD-R HTL is a recordable Blu-ray type whose written marks have lower optical reflectivity than the unrecorded material.
- · BD-R LTH
BD-R LTH is an organic-dye recordable Blu-ray type whose written marks have higher optical reflectivity than the unrecorded material.
- · Blu-ray data archive
A Blu-ray data archive stores retained digital content on recordable BD-R or rewritable BD-RE optical media.
- · Btrfs
Btrfs is a Linux filesystem built around copy-on-write B-trees, with integrated snapshots, checksumming and multi-device storage.
- · Btrfs read-only subvolume
A Btrfs read-only subvolume is a subvolume whose ro property prevents modification of its file contents and directory hierarchy.
- · Btrfs tree log
The Btrfs tree log is a per-subvolume recovery structure that can persist file or directory changes for fsync without a full filesystem transaction commit.
- · CD block error rate (BLER)
CD BLER counts blocks containing detectable errors per second at the input to the C1 decoder.
- · Compensation log record
An ARIES compensation log record describes a rollback action for redo and records where undo processing should continue.
- · Conditional write
A conditional write changes a resource only when a server-evaluated precondition, such as a matching entity tag or the absence of a current object, holds.
- · Consumer SSD
An SSD intended for personal computers and other client devices rather than shared server workloads.
- · copy_file_range
copy_file_range is a Linux system call that copies a requested file range in the kernel and reports the number of bytes actually copied.
- · Copy-on-write (storage)
A technique that writes modified blocks to new locations instead of overwriting in place.
- · Crash consistency
Crash consistency is the property that persistent state remains within an application's allowed recovery states after an unexpected interruption.
- · Cryptographic nonce
A cryptographic nonce is an input value whose required uniqueness scope is specified by the cryptographic construction that uses it.
- · Degraded array
A degraded array is a redundant storage array operating with less than its configured healthy redundancy because one or more members are unavailable or not fully synchronized.
- · Delete marker
A delete marker is a payload-free S3 version that makes a key appear deleted when the marker is current.
- · Device-managed SMR (DM-SMR)
An SMR HDD that presents a conventional LBA interface while the firmware relocates writes across shingled bands.
- · Directory fsync
Directory fsync is synchronization of a directory descriptor to persist directory state, including changes to the names through which files are reached.
- · Discard
A block-storage request indicating that specified logical ranges are no longer needed by the caller.
- · dm-crypt
dm-crypt is a Linux device-mapper target that transparently encrypts and decrypts block-device I/O using the kernel cryptographic API.
- · dm-integrity
dm-integrity is a Linux device-mapper target that stores per-block integrity tags and can generate and verify them when configured with an internal hash.
- · dm-integrity recovery mode
dm-integrity recovery mode is a Linux device-mapper access mode that bypasses integrity checks and journal replay while disallowing writes through the mapping.
- · dm-verity
dm-verity is a Linux device-mapper target that verifies a read-only block mapping against a trusted cryptographic root hash.
- · DVD PI Sum8
DVD PI Sum8 counts parity-inner rows containing errors over eight consecutive ECC blocks before correction.
- · Enterprise SSD
An enterprise SSD is a solid-state drive intended for server, storage-array or other multi-user storage environments.
- · Entity tag
An entity tag is an opaque HTTP validator assigned by a server to a selected representation for cache validation and conditional requests.
- · Erasure coding
Erasure coding encodes data with redundancy so that missing fragments at known locations can be reconstructed from a sufficient, code-dependent set of surviving fragments.
- · Error-and-erasure decoding
Error-and-erasure decoding reconstructs a codeword while accounting for both known erased positions and incorrect values at unknown positions.
- · errseq_t
errseq_t is a Linux kernel datatype that records an error and sequence state so observers can detect failures since their saved cursor.
- · Ext4 fast commit
Ext4 fast commit is a journaling mechanism that records metadata deltas for eligible updates instead of a full traditional metadata-block commit.
- · Ext4 orphan file
An ext4 orphan file is an internal metadata file that stores inode-number slots for orphan cleanup when the orphan_file filesystem feature is enabled.
- · Ext4 orphan inode
An ext4 orphan inode is an inode tracked so recovery can finish deletion or remove excess allocations left by an interrupted truncation or extension.
- · F2FS
F2FS is a Linux log-structured filesystem designed for flash storage devices that expose a block interface.
- · F2FS checkpoint
An F2FS checkpoint records a consistent filesystem recovery point, including information selecting valid metadata-table versions.
- · F2FS roll-forward recovery
F2FS roll-forward recovery reconstructs selected post-checkpoint file updates from recovery-marked node blocks.
- · Fail-slow
Fail-slow describes a component that remains operational but performs more slowly than expected for its operating conditions.
- · fdatasync
fdatasync is a file-synchronization operation that persists modified data and the metadata required to retrieve it, without requiring synchronization of all other file metadata.
- · Filesystem freeze
A filesystem freeze temporarily blocks modifications after completing ongoing filesystem transactions and flushing dirty filesystem data, metadata and logs to storage.
- · Flash translation layer (FTL)
A flash translation layer maps host logical addresses to physical flash locations and manages flash-specific operations such as reclamation.
- · fs-verity
fs-verity is a Linux filesystem facility that verifies the immutable contents of selected files against a cryptographic file digest.
- · Fuzzy checkpoint
A fuzzy checkpoint records recovery information while transaction processing continues, rather than freezing all activity into one instantaneous database image.
- · Garbage collection (flash)
Flash garbage collection reclaims a victim erase block by first preserving its valid data elsewhere and then erasing the block for reuse.
- · Ghost cache entry
A ghost cache entry records the identity or history of an evicted item without retaining its cached payload.
- · Group commit
Group commit is a logging optimization that shares a durable log synchronization across multiple concurrent transaction or write requests.
- · Hard disk drive (HDD)
A non-volatile storage device that records data on spinning platters with a moving read/write head.
- · Head slap
Head slap is a head-disk contact mechanism in which an HDD slider separates from the disk and then snaps back into it.
- · Hot spare
A hot spare is a standby device or reserved distributed spare capacity available to take over a failed member's role in a redundant array or pool.
- · Idempotency
Idempotency is the property that repeating the same operation with identical inputs has the same intended effect as performing it once.
- · Idempotency key
An idempotency key identifies attempts of one logical request so a service can recognize retries within its documented scope and retention rules.
- · Inode
An inode is a filesystem object that holds a file's metadata and has a filesystem-local number, distinct from the directory names and open file descriptions that refer to the file.
- · io_uring integrity metadata
io_uring integrity metadata is a separately described protection-information buffer attached to a read or write request through SQE attributes.
- · Journal checkpoint
In JBD2 terminology, a journal checkpoint writes committed journaled state to its home locations so the corresponding log space can be reclaimed.
- · Legal hold (storage)
A storage legal hold prevents deletion or replacement of protected data until an authorized action removes the hold, rather than expiring on a preset date.
- · Linux block-device read-only flag
The Linux block-device read-only flag is a software-controlled device state exposed through block-device ioctls to restrict write access.
- · Logical journaling
Logical journaling records recoverable changes to logical objects or operations rather than simply reproducing their on-disk representation.
- · LSM compaction
LSM compaction reorganizes sorted storage components, commonly merging selected runs or SST files and removing obsolete entries when visibility rules permit.
- · LTFS
Linear Tape File System is a self-describing tape format that stores files and directory metadata for interchange between compatible implementations.
- · LTFS data placement policy
An LTFS data placement policy specifies which file data may be stored in the index partition alongside filesystem metadata.
- · LTFS index
An LTFS index is XML metadata describing files, directories and the mapping from file contents to recorded tape extents.
- · LUKS
LUKS is a disk-encryption format that records cryptographic parameters and key-management metadata, including keyslots used to unlock encrypted storage.
- · M-DISC
M-DISC is a family of write-once optical media marketed for archival storage using an inorganic recording layer.
- · Maximum distance separable code
A maximum distance separable code achieves minimum distance n - k + 1 for length n and dimension k, allowing recovery from any n - k known symbol erasures.
- · Mean time to data loss (MTTDL)
Mean time to data loss (MTTDL) is the expected time until a storage system first loses data that cannot be retrieved or reconstructed.
- · Medium Auxiliary Memory
Medium Auxiliary Memory is nonvolatile storage associated with an individual storage medium, used to hold auxiliary attributes about the medium and its recorded data.
- · Memtable
A memtable is an in-memory structure that holds recent LSM-tree updates before they are flushed to sorted table files.
- · Metadata update dependency
A metadata update dependency is an ordering requirement between persistent changes needed to preserve a filesystem invariant.
- · Mirror set
A mirror set is a group of storage members configured to maintain copies of the same logical data.
- · Mirroring (storage)
Storage mirroring maintains a copy of source data on another storage device or system.
- · NAND erase block
A NAND erase block is a group of flash pages erased together by a block-erase operation.
- · NAND page
A NAND page is a group of data bits addressed together for a page read or program operation within an erase block.
- · NAND wordline
A NAND wordline is a shared control-gate connection used to address a group of flash memory cells.
- · Native Command Queuing
Native Command Queuing (NCQ) allows multiple tagged ATA commands to be outstanding so a supporting device can schedule their execution.
- · Normalized magnitude of data loss (NOMDL)
Normalized magnitude of data loss (NOMDL) is the expected number of bytes lost per usable terabyte within a stated mission time.
- · NVMe Available Spare
NVMe Available Spare is the normalized percentage of spare capacity remaining in the NVM subsystem.
- · NVMe Data Units Written
NVMe Data Units Written is the SMART / Health Information log counter of host-written data, excluding metadata, in rounded-up groups of 1,000 units of 512 bytes.
- · NVMe Device Self-test
NVMe Device Self-test is an optional controller-executed diagnostic capability with short and extended operations whose namespace scope is selected by the initiating command.
- · NVMe Device Self-test log
The NVMe Device Self-test log is log page 06h, which reports current operation progress and twenty slots for results of completed or aborted self-test operations.
- · NVMe Endurance Group
An NVMe Endurance Group is a scope within which an NVM subsystem manages endurance.
- · NVMe Error Information Log Entries
NVMe Error Information Log Entries is the SMART / Health Information log counter of Error Information log entries accumulated over the controller's life.
- · NVMe multipath I/O policy
An NVMe multipath I/O policy is a host-side rule for choosing among eligible routes to one namespace.
- · NVMe Percentage Used
NVMe Percentage Used is a vendor-specific estimate of the percentage of NVM endurance consumed by actual use.
- · O_SYNC
O_SYNC is an open-file status flag requiring write operations to complete with synchronized I/O file integrity, covering data and associated file metadata.
- · Object lock
An object lock enforces retention restrictions on a specified object version through a retention policy or hold.
- · Object storage
Object storage organizes data as independently addressed objects with associated metadata, accessed through object identifiers or keys rather than a block-device interface.
- · Object version ID
An object version ID is a service-assigned identifier used with a bucket and key to select an S3 object version.
- · Object versioning
Object versioning retains separately identifiable revisions of an object under the same key.
- · Open-channel SSD
An open-channel SSD exposes media parallelism and placement controls so host software can manage data placement and I/O scheduling.
- · Optical archival storage
Optical archival storage preserves data on removable laser-readable media as part of a long-term retention and migration workflow.
- · Optical-disc lifetime estimate
An optical-disc lifetime estimate predicts time to a defined error threshold for a stated population percentile and storage environment under an aging model.
- · Page LSN
A page LSN records the log sequence number of the latest logged change reflected in a database page under the engine's recovery protocol.
- · Parity RAID
Parity RAID uses computed redundancy across striped data to reconstruct missing chunks without storing complete mirrored copies.
- · Persistence ordering
Persistence ordering is a constraint on the order in which updates become durable, rather than merely the order in which an application issues them.
- · Physical journaling
Physical journaling records modified on-disk representations, such as metadata buffer contents or changed regions, for recovery.
- · POSIX_FADV_DONTNEED
POSIX_FADV_DONTNEED is a file-access advice value requesting release of cached data that the application does not expect to access soon.
- · POSIX_FADV_NOREUSE
POSIX_FADV_NOREUSE is a file-access advice value indicating that the application does not expect to reuse the specified data.
- · Prefix-preserving crash consistency
A recovery contract in which surviving updates form a prefix of the specified write order, including all writes covered by the last successful flush.
- · Protection information (PI)
Protection information is integrity metadata accompanying storage I/O, with checks determined by the protocol, protection format and participating layers.
- · RAID 10
RAID 10 combines replicated data copies with striping across storage members.
- · RAID 5
RAID 5 stripes data and distributed single parity across members so that one missing member can be reconstructed when the remaining required information is readable and consistent.
- · RAID 6
RAID 6 uses dual parity protection to retain data after any two member failures, provided the remaining required data and redundancy are readable and consistent.
- · RAID rebuild
A RAID rebuild reconstructs the required contents of a replacement or out-of-sync member from the array's surviving redundancy.
- · RAID tail hour
In The Tail at Store, a RAID tail hour is a group-hour in which at least one data drive has an hourly average I/O latency at least twice its group's median.
- · Read-only mount
A read-only mount restricts filesystem modification through Linux mount state, whose scope can be a VFS mount node, the filesystem superblock, or both.
- · Reed–Solomon codes (storage)
Reed–Solomon codes are maximum-distance-separable error-correcting codes over finite-field symbols that add n − k check symbols to k data symbols and can rebuild missing (erased) symbols, which is why storage systems use them for RAID 6 parity and erasure-coded object storage.
- · Replay attack
A replay attack reuses previously valid data to induce acceptance in an unauthorized context.
- · Retry deduplication
Retry deduplication recognizes repeated submissions of one logical request so that an already completed mutation is not executed again.
- · RocksDB pending compaction bytes
RocksDB pending compaction bytes is an estimate of rewrite work needed to bring LSM levels within their size targets under level-based compaction.
- · RocksDB write slowdown
A RocksDB write slowdown is the rate-reduction state of write backpressure, distinct from a full write stop.
- · RocksDB write stall
A RocksDB write stall is backpressure that delays or stops foreground writes when configured resource or background-work limits are reached.
- · RocksDB WriteBufferManager
RocksDB WriteBufferManager is a shared controller that accounts for memtable memory across participating column families or database instances and can trigger flushing, optional write stalls and block-cache charges.
- · Rollback journal
A rollback journal is an undo log that preserves original database content so an incomplete transaction can be reversed.
- · Rollback protection
Rollback protection rejects protected state older than an enforced version or freshness boundary using trusted state that cannot be rolled back with the protected object.
- · Rotational vibration (RV)
Rotational vibration is angular oscillatory motion that can disturb an HDD's ability to keep its heads aligned with recording tracks.
- · RWF_DONTCACHE
RWF_DONTCACHE is a Linux per-operation flag requesting best-effort removal of page-cache content newly instantiated by a buffered read or write.
- · SATA PHY event counters
SATA PHY event counters record distinct interface events in General Purpose Log 0x11 rather than measuring a disk failure rate.
- · Serial ATA (SATA)
A serial bus interface and protocol family for attaching HDDs and SSDs, widely used in PCs and servers.
- · Shingled band
A shingled band is a group of overlapping magnetic tracks separated from neighboring bands by guard regions that limit overwrite propagation.
- · Slowdown ratio
In Perseus, the slowdown ratio is observed drive latency divided by the workload-dependent upper prediction bound for that latency.
- · smartctl exit status
The smartctl exit status is a process result whose normal ATA interpretation combines independent command, health-threshold and log-history flags in an eight-bit mask.
- · SMR garbage collection
SMR garbage collection relocates staged updates and rewrites affected shingled regions to reclaim reusable space.
- · SMR persistent cache
An SMR persistent cache is a nonvolatile staging area that holds updates before they are incorporated into their target shingled regions.
- · Snapshot-consistent flash translation layer
A research FTL that recovers a flush-defined disk snapshot instead of an arbitrary later write prefix.
- · Soft updates
Soft updates is a filesystem metadata-consistency mechanism that delays unsafe on-disk changes until their update dependencies are satisfied.
- · Solid-state drive (SSD)
A solid-state drive is a storage device that retains data in nonvolatile semiconductor memory without moving recording media.
- · SSD mapping cache
An SSD mapping cache retains selected logical-to-physical address mappings in memory.
- · Storage immutability
Storage immutability prevents specified data from being overwritten or deleted under an enforced retention rule or hold.
- · Storage latency
Storage latency is the elapsed time between specified start and completion events for an I/O operation, including any delays inside those measurement boundaries.
- · Storage snapshot
A storage snapshot is a point-in-time representation of a filesystem or volume, whose implementation may share blocks with its source.
- · Symbol erasure
A symbol erasure is a codeword position whose value is unavailable or deliberately discarded and whose location is known to the decoder.
- · TapeAlert
TapeAlert is a SCSI status-reporting mechanism for tape devices that exposes condition flags through LOG SENSE page 2Eh.
- · Thermal flying-height control (TFC)
Thermal flying-height control uses an embedded heater to adjust HDD head-disk clearance through local thermal protrusion.
- · Touchdown sensor (TDS)
An HDD touchdown sensor is a temperature-sensitive element used to observe thermal changes near the head-disk interface during clearance and contact studies.
- · UASP
USB Attached SCSI Protocol (UASP) transports SCSI commands over USB and supports multiple outstanding commands, without by itself guaranteeing throughput, SMART pass-through or durable writes.
- · USB mass storage
The USB device class that presents block storage to hosts using BOT or UASP transports.
- · WORM storage
Write-once-read-many storage preserves recorded content against subsequent alteration while allowing repeated reads.
- · Writeback error
A writeback error is a failure encountered while transferring modified cached data to its backing storage.
- · XFS Committed Item List
The XFS Committed Item List tracks in-memory logged changes awaiting a journal checkpoint.
- · XTS-AES
XTS-AES is a tweakable AES encryption mode for storage data units that provides confidentiality without authenticating the data or its source.
- · Active-zone limit
An active-zone limit caps the zones simultaneously in implicitly open, explicitly open or closed states in a zoned namespace.
- · Adaptive Read-Retry
Adaptive Read-Retry selects shorter NAND precharge timing from profiled conditions while reserving error-correction margin.
- · Backup
A backup is a retained copy of data or software intended to support later recovery.
- · Backup manifest
A backup manifest is a machine-readable inventory of expected backup contents and associated verification metadata.
- · Checkpoint starvation
Checkpoint starvation is sustained inability to finish or reset a database log checkpoint because the required progress conditions keep being obstructed.
- · Class prevalence
Class prevalence is the proportion of observations belonging to a specified class in a defined population or evaluation sample.
- · Clear sanitization
Clear sanitization uses logical operations across all user-addressable storage locations to resist simple, non-invasive recovery through the device interface available to the user, such as the host interface.
- · Close-to-open consistency
Close-to-open consistency is a file-sharing cache policy that combines writeback at close with validation at a later open to support visibility of updates between sequential users.
- · Cryptographic erase
Cryptographic erase sanitizes the keys protecting encrypted target data so that recovering the plaintext becomes infeasible.
- · Data sanitization
Media sanitization is a process that makes recovery of target data infeasible for a specified level of effort.
- · Database checkpoint
A database checkpoint records engine-defined recovery state to bound or reduce the work needed after a crash.
- · Database write amplification
Database byte write amplification is storage-engine output bytes written divided by logical input bytes over a stated interval and accounting boundary.
- · Destroy sanitization
Destroy sanitization makes target-data recovery infeasible even with current state-of-the-art laboratory techniques and leaves the medium incapable of storing data again.
- · Differential backup
A differential backup captures changes accumulated since its designated full-backup base, rather than only changes since the preceding differential backup.
- · Direct I/O
Direct I/O is a file-access mode, such as Linux O_DIRECT, that requests I/O through user-space buffers while minimizing use of the local operating-system page cache.
- · Direct-I/O alignment
Direct-I/O alignment is the per-file constraint on memory addresses, file offsets and I/O segment lengths used for direct transfers.
- · Dynamic layout score
Dynamic layout score is a workload-specific locality measure based on the fraction of traced filesystem block requests that are contiguous in logical block address space.
- · Fault injection
Fault injection is a testing method that deliberately introduces specified faults into a system to observe detection, propagation and recovery under a defined workload.
- · Filesystem aging
Filesystem aging is the evolution of allocation layout through file operations that can degrade performance even when the logical contents remain comparable.
- · Free-space fragmentation
Free-space fragmentation is the division of a filesystem's available allocation space into separated contiguous runs.
- · Full backup
A full backup captures every data object within a defined backup scope, whether or not it has changed since an earlier backup.
- · Grouped cross-validation
Grouped cross-validation evaluates a predictor using splits that keep all observations from each specified group together in either training or testing within a split.
- · Hot journal
A hot journal is a SQLite rollback journal that requires replay to recover an interrupted database transaction.
- · Incremental backup
An incremental backup records changes relative to an earlier backup state and reuses retained backup data to represent unchanged content.
- · io_uring completion queue entry
An io_uring completion queue entry is a result record containing an operation result, flags and application-provided identification data.
- · io_uring drain
io_uring drain is a submission flag that delays a request until earlier requests in the same ring finish and delays later requests until the drain request finishes.
- · Journal revoke record
A journal revoke record identifies a block whose older journaled images must not be replayed over newer uses of that block.
- · Leveled compaction
Leveled compaction organizes sorted data into levels with size targets and merges selected data with overlapping key ranges in a destination level.
- · NAND read-reference voltage
A NAND read-reference voltage is a sensing voltage used to distinguish stored cell states during a read operation.
- · NAND read-retry
NAND read-retry rereads a flash page with adjusted sensing settings after an unsuccessful error-correction attempt.
- · NFS attribute cache
An NFS attribute cache is a client-side copy of file or directory metadata reused until refresh or revalidation is required.
- · NFS lookup cache
An NFS lookup cache retains directory-name resolution results, including found and missing entries, subject to revalidation.
- · NFSv4 file delegation
An NFSv4 file delegation is a recallable server grant allowing a client to manage specified opens and cached file state locally while conflicting access is coordinated by the server.
- · NVMe Controller Busy Time
NVMe Controller Busy Time is the cumulative number of minutes during which the controller has an outstanding I/O command, as defined by the SMART / Health Information log.
- · NVMe Media and Data Integrity Errors
NVMe Media and Data Integrity Errors is the SMART / Health Information log counter of unrecovered data-integrity errors detected by the controller.
- · NVMe sanitize
NVMe Sanitize starts a controller-managed sanitization operation using a supported Block Erase, Overwrite or Crypto Erase action.
- · NVMe Unexpected Power Losses
NVMe Unexpected Power Losses is the SMART / Health Information log counter of main-power losses that meet the specification's incomplete-shutdown conditions.
- · Open-zone limit
An open-zone limit caps the zones simultaneously in implicitly open or explicitly open states.
- · Partial quorum
A partial quorum configuration allows a read response set and a write acknowledgement set that need not overlap.
- · Pipelined Read-Retry
Pipelined Read-Retry overlaps NAND retry sensing with transfer and decoding of earlier attempts.
- · Point-in-time recovery
Point-in-time recovery restores a database to a selected historical state by applying retained transaction-log history to a suitable backup.
- · Powersafe overwrite
Powersafe overwrite is the storage property that an interrupted file write leaves bytes outside its requested range unchanged.
- · Purge sanitization
Purge sanitization uses suitable logical or physical techniques to make target-data recovery infeasible even with current state-of-the-art laboratory techniques while potentially allowing the medium to be reused.
- · Quorum intersection
Quorum intersection is the property that every relevant pair of allowed quorums shares at least one replica.
- · Read amplification
Read amplification expresses storage-read work per logical read request or per useful byte returned, under an explicitly stated accounting convention.
- · Read disturb
Read disturb is an unintended change in NAND cell thresholds caused by reads to other wordlines in the same block.
- · Read reclaim
Read reclaim relocates valid NAND data to fresh pages to limit accumulated read-disturb errors.
- · Read-pass voltage
Read-pass voltage is the bias applied to unselected NAND wordlines to make their cells conduct while another wordline is sensed.
- · Recovery point objective
The recovery point objective specifies the point in time to which data must be recoverable after a disruption, commonly expressed as the maximum tolerable data-loss interval.
- · Recovery target
A recovery target is the configured condition at which database log replay should stop.
- · Restore test
A restore test exercises recovery from backup material and checks whether the recovered data or selected system functions meet defined acceptance criteria.
- · RocksDB atomic flush
RocksDB atomic flush installs the results of flushing selected column families into the database as one logical all-or-nothing MANIFEST group.
- · RocksDB MANIFEST
RocksDB MANIFEST is the version-edit log mechanism used to recover the database's recorded state, with CURRENT identifying the active manifest log.
- · Sanitization validation
Sanitization validation decides whether to accept or reject a sanitization result after evaluating its evidence, suitability and remaining risk to data confidentiality.
- · Sanitization verification
Sanitization verification inspects the outcome of a sanitization operation to determine whether the selected technique completed successfully.
- · Sloppy quorum
A sloppy quorum allows reachable fallback nodes outside a key's normal replica set to participate when intended replicas are unavailable.
- · SMB byte-range lock
An SMB byte-range lock coordinates access to a specified region of a file through shared or exclusive locking associated with a file open.
- · SMB file sharing
Server Message Block is a network protocol family used to access shared files and other server resources.
- · SMB FLUSH
SMB FLUSH is an SMB2/SMB3 request to propagate cached information for an open file to its backing persistent store, or to wait for a named-pipe reader to consume previously written data.
- · SMB lease
An SMB lease is a server-granted arrangement that permits selected client-side read, write, or handle caching while coordinating access to a shared object.
- · SMB opportunistic lock
An SMB opportunistic lock is a revocable server-granted caching arrangement that lets a client reduce network operations while coordinating access to a file.
- · SMB persistent handle
An SMB persistent handle extends durable-open recovery to server failover on a supporting continuously available share.
- · Sorted string table
An immutable table of sorted key-value entries used as a persistent component in storage engines such as LevelDB and RocksDB.
- · Space amplification
Database space amplification quantifies storage overhead relative to logical data size under an explicitly stated total-space or excess-space convention.
- · SQLite super-journal
A SQLite super-journal is a coordination file that links recovery outcomes for a qualifying rollback-mode transaction modifying multiple attached databases on one connection.
- · SSD preconditioning
Preparatory writing used to establish a specified SSD state before performance measurement.
- · SSD steady state
A workload-specific SSD performance condition in which a named metric satisfies stated stability criteria over a defined measurement window.
- · Synthetic full backup
A synthetic full backup reconstructs a complete backup for a selected recovery point from retained full and incremental backup data without rereading the entire protected dataset from production.
- · Temporal validation
Temporal validation evaluates a predictor on later observations using training information available before the evaluation boundary.
- · Tiered compaction
Tiered compaction allows multiple sorted runs to accumulate and merges groups of runs into typically larger runs instead of maintaining one run at every tier.
- · WAL archiving
WAL archiving preserves write-ahead-log segments for later recovery before normal log recycling can remove them.
- · ZFS Intent Log
The ZFS Intent Log is a per-dataset recovery log that can persist filesystem operations before their transaction group reaches stable storage.
- · ZFS separate log
A ZFS separate log is dedicated pool log-device storage used for intent-log blocks instead of their default allocation in the main pool.
- · ZFS transaction group
A ZFS transaction group is an ordered batch of changes whose synchronization advances the pool's consistent persistent state.
- · Acting set (Ceph)
In Ceph, the acting set is the ordered OSD list responsible for a placement group at a particular map epoch.
- · Additional sense code
The additional sense code (ASC) is an eight-bit SCSI sense-data field that refines the sense key and is interpreted together with its qualifier.
- · Additional sense code qualifier
The additional sense code qualifier (ASCQ) is an eight-bit SCSI sense-data field interpreted with the ASC to identify a more specific condition.
- · Asynchronous DRAM Refresh
Asynchronous DRAM Refresh is a platform power-fail mechanism that drains covered memory-controller write-pending buffers to persistent memory without protecting dirty CPU caches.
- · Btrfs NODATASUM
Btrfs NODATASUM marks file data as excluded from the filesystem's data checksumming.
- · Byte-addressable storage
Byte-addressable storage exposes individual byte locations rather than only block-sized addresses, without by itself specifying atomicity or durability.
- · Completion latency (fio)
Completion latency in fio is the engine-dependent interval from submission of an I/O unit to observation of its completion.
- · CRUSH
CRUSH is a deterministic, pseudorandom placement algorithm that maps an identifier to storage targets using a weighted cluster hierarchy and placement rules.
- · Differential observability
Differential observability is a difference between entities' observations of the same system's health.
- · Direct Access
Direct Access is a Linux mechanism that lets suitable memory-like storage serve file data without an intervening page-cache copy.
- · Distributed snapshot
A distributed snapshot records a coordinated global state of participating processes and their communication channels.
- · Distributed spare
A distributed spare is replacement capacity spread across multiple devices rather than reserved on a single dedicated spare device.
- · Extended Asynchronous DRAM Refresh
Extended Asynchronous DRAM Refresh is a platform capability that extends persistent-memory power-fail protection to CPU caches.
- · Filesystem consistency invariant
A filesystem consistency invariant is a rule relating filesystem metadata that must hold at a defined consistency boundary.
- · Gray failure
A gray failure occurs when an application observes a system as unhealthy while that system's failure observer regards it as healthy.
- · Hinted handoff
Hinted handoff retains information about an update missed by an intended replica and attempts to deliver it later.
- · Hole punching
Deallocating whole filesystem blocks within a file range while preserving the file's logical length and making the affected range read as zeros.
- · I/O request size
I/O request size is the number of bytes requested by one input/output operation at a specified system boundary.
- · INCITS T10 SCSI
The standards body maintaining SCSI command sets used by SAS, FC, and iSCSI block devices.
- · IOPS
Observed IOPS is the number of completed input/output operations divided by the measurement duration at a specified system boundary.
- · Left-truncation
Left-truncation is selection into a time-to-event sample only after a unit has survived to its observation-entry time.
- · Linearizability
Linearizability requires concurrent operations on an object to behave as a legal sequential execution that preserves the real-time order of non-overlapping operations.
- · Little's law
Little's law relates time-averaged occupancy L to the operation rate lambda and mean residence time W within the same system boundary.
- · LSM tombstone
An LSM tombstone is a versioned deletion marker that suppresses older values for a key without immediately erasing their stored bytes.
- · MAP_SYNC
MAP_SYNC is a Linux mmap flag for supported DAX files that provides a file-offset persistence guarantee used with appropriate CPU persistence instructions.
- · Monotonic reads
Monotonic reads is a session guarantee requiring successive reads to observe states that include the effects of writes relevant to the session's earlier reads.
- · Monotonic writes
Monotonic writes requires every replica incorporating a later session write to incorporate the session's earlier writes and order them before it.
- · NAND cross-temperature effect
The NAND cross-temperature effect is the change in read behavior associated with programming and subsequently reading a cell at different temperatures.
- · Parity declustering
Parity declustering places parity groups across a larger device pool to distribute reconstruction work.
- · Peering (Ceph)
Ceph peering establishes agreement among OSDs about a placement group's object state and associated metadata.
- · Persistence domain
A persistence domain is the set of components whose stored contents are guaranteed to survive a specified failure under the platform's operating contract.
- · Persistent memory
Persistent memory is nonvolatile storage exposed through a memory-addressable interface, with durability governed by the platform's persistence contract.
- · Placement group (Ceph)
In Ceph, a placement group is a logical grouping of objects within a pool that share storage placement and are managed together for replication and recovery.
- · Prefix Bloom filter
A Bloom filter built over extracted key prefixes rather than only whole keys, allowing negative membership tests for those represented prefixes.
- · Prefix extractor
A key transformation with a defined input domain that produces the prefix used for prefix-based indexing or filtering.
- · Queue depth
Queue depth is the number of outstanding I/O requests at a specified boundary, while a configured depth limit is the maximum or target allowed in flight.
- · Read quorum
A read quorum is a protocol-approved set of replicas whose responses satisfy the requirement for completing a read.
- · Read repair
Read repair reconciles divergent replicas as part of processing a read request.
- · Read-your-writes
Read-your-writes is a session guarantee requiring subsequent reads to use a state that incorporates the session's prior successful writes.
- · Recovery time objective
The recovery time objective sets the tolerable duration of resource unavailability before its effect on supported processes becomes unacceptable.
- · Resilver
A resilver is ZFS's process of bringing a newly attached, replacement or out-of-date device into agreement with the data it must hold.
- · SCSI
Small Computer System Interface (SCSI) is a family of standards for device commands, logical units, status reporting and the transports that carry them.
- · SCSI CHECK CONDITION
CHECK CONDITION is a SCSI command status used to report a condition described by sense information.
- · SCSI sense key
A SCSI sense key is a four-bit field identifying the broad category of a condition reported in sense data.
- · Session guarantees
Session guarantees are consistency contracts for a session's operations, including read-your-writes, monotonic reads, monotonic writes and writes-follow-reads.
- · SMB durable handle
An SMB durable handle allows a client to attempt to recover an existing file open after a network disconnect.
- · SMOTE
SMOTE is an oversampling method that creates synthetic minority-class feature vectors by interpolating between a minority example and selected minority-class nearest neighbors.
- · Storage throughput
Storage throughput is the amount of data transferred per unit time at a specified system boundary, usually expressed in bytes per second.
- · Submission latency (fio)
Submission latency in fio is the elapsed time from initialization of an I/O unit to an engine-dependent submission boundary.
- · Succinct range filter
An approximate point-and-range membership structure based on a truncated order-preserving Fast Succinct Trie.
- · Survival function
The survival function S(t) = P(T > t) is the probability that a time T to a defined event exceeds t, measured from a specified time origin.
- · sync_file_range
sync_file_range is a Linux-specific operation that initiates or waits for page-cache writeback over a file range without synchronizing file metadata or flushing volatile device write caches.
- · Systematic code
A systematic code includes the original data symbols unchanged among its encoded symbols.
- · Thin provisioning
Presenting logical storage capacity while allocating physical backing from a shared pool as writes require it.
- · Unserializable write
An unserializable write is a storage failure in which recovered contents cannot be reconciled with an operation order allowed by the known write history.
- · Up set (Ceph)
In Ceph, the up set is the ordered OSD list assigned to a placement group by CRUSH for a particular map epoch.
- · Write quorum
A write quorum is a protocol-approved set of replicas whose acknowledgements satisfy the requirement for completing a write.
- · Writes-follow-reads
Writes-follow-reads requires every replica incorporating a session write to incorporate and order before it the updates relevant to that session's earlier reads.
- · Zone capacity
Zone capacity is the maximum number of logical blocks that may be written within a zone.
- · Zone Finish
Zone Finish changes an eligible sequential zone to Full, ending further writes until reset without discarding the data already written.
- · Zone Reset
Zone Reset makes an eligible sequential zone empty for reuse, returns its write pointer to the zone start and invalidates access to its previous data.
- · Zone write pointer
A zone write pointer identifies the next writable logical block address in a sequential zone when its current state makes that pointer valid.
- · Zoned namespaces (ZNS)
Zoned Namespaces (ZNS) is an NVMe command set that organizes a namespace into zones with sequential-write constraints and explicit reset before reuse.
- · 4Kn native sectors
4Kn exposes 4096-byte logical sectors and 4096-byte physical sectors without 512-byte sector emulation.
- · 512e emulation
512e exposes 512-byte logical sectors over 4096-byte physical sectors.
- · Active LBA range
The specified logical-block address interval available to a storage test or its preparation workload.
- · Advanced Format (4K)
Advanced Format is a disk-sector architecture expanded beyond legacy 512-byte sectors, with 512e and 4Kn as its familiar 4K variants.
- · Annual replacement rate
Annual replacement rate (ARR) counts recorded device replacements relative to device exposure, scaled to one year.
- · Bit error rate
Bit error rate (BER) expresses erroneous bits relative to the total bits considered at a specified observation point.
- · Chain replication
Chain replication orders replicas so that updates enter at a head, propagate along the chain and complete at a tail that also serves queries.
- · Codeword (storage ECC)
A storage ECC codeword is an encoded unit representing a data chunk together with error-correction redundancy.
- · Compare-and-swap
Compare-and-swap is an atomic conditional update that replaces a stored value only if it matches an expected value and reports the comparison outcome or observed value.
- · Cryptographic hash function
A cryptographic hash function deterministically produces a digest and is designed to resist preimage, second-preimage and collision attacks.
- · Data deduplication
Data deduplication replaces duplicate content with references to shared stored blocks or chunks within a defined matching scope.
- · Delete persistence latency
In delete-aware LSM research, delete persistence latency is the delay from inserting a tombstone to removing the entries it invalidates through compaction.
- · Error-correcting code
An error-correcting code adds structured redundancy to data so a decoder can detect and correct specified error patterns.
- · Force Unit Access
Force Unit Access is a write-request property requiring the flagged data to reach nonvolatile storage before that request is reported complete.
- · FTL translation page
An FTL translation page is a flash page containing address-mapping entries.
- · Greedy garbage collection
A flash block-cleaning policy that selects a candidate erase block with the fewest valid pages, minimizing the valid data copied for that selection.
- · Hash collision
A hash collision occurs when two distinct inputs produce the same digest under the same hash function and parameters.
- · Host memory buffer (HMB)
Host Memory Buffer is an optional NVMe feature through which a controller uses assigned host-memory ranges.
- · Hot/cold data separation
A flash-placement strategy that groups data by update frequency so frequently overwritten pages and long-lived pages need not share the same cleaning pool.
- · Key-value separation
Key-value separation stores values outside the main key index and retains references in that index.
- · LDPC (NAND ECC)
An LDPC code is an error-correcting linear block code defined by a sparse parity-check matrix.
- · Logical sector
A logical sector is the addressable data unit selected by a logical block address at a storage interface.
- · MTBF
Mean time between failures (MTBF) is the expected operating time between successive failures of a repairable system under specified conditions.
- · NAND channel
A NAND channel is a communication path between an SSD controller and flash components sharing its transfer resources.
- · NFS write verifier
An NFSv3 write verifier is a server cookie returned by WRITE and COMMIT that lets clients detect state changes that may have lost uncommitted writes.
- · NVMe FDP
An optional NVMe capability that lets a host guide write placement through placement identifiers while retaining conventional logical-block reads and writes.
- · NVMe placement identifier
An FDP value combining a reclaim group identifier with a namespace-scoped placement handle that maps to a reclaim unit handle.
- · NVMe reclaim unit
An FDP storage abstraction within a reclaim group that the controller can physically erase independently of other reclaim units.
- · NVMe reclaim unit handle
An FDP controller resource that references one reclaim unit in each reclaim group of an endurance group.
- · Over-provisioning
SSD over-provisioning is flash capacity withheld from host-addressable user storage to support controller management such as garbage collection and wear leveling.
- · Partition alignment
Partition alignment places a partition's starting offset on boundaries appropriate to the storage topology reported by the underlying stack.
- · Physical sector
A physical sector is the device-level data unit whose reported size describes physical read/write granularity rather than logical addressing.
- · RAID
Redundant Array of Independent Disks combines multiple drives for performance or fault tolerance.
- · Range tombstone
A range tombstone is a versioned deletion marker that suppresses covered older keys across an ordered key interval.
- · Raw bit error rate
Raw bit error rate is the number of erroneous bits divided by the total bits read at a specified measurement point before error correction.
- · Rolling hash
A rolling hash updates a fingerprint incrementally as the input window or stream advances, instead of recomputing it over the entire input each time.
- · RWF_ATOMIC
RWF_ATOMIC is a Linux pwritev2 flag requesting torn-write protection for the data range of a supported direct-I/O write.
- · Serializability
Serializability is a transaction-isolation property under which the read results and writes of committed concurrent transactions are equivalent to some serial execution of those transactions.
- · Snapshot isolation
Snapshot isolation gives a transaction a stable committed snapshot plus its own writes and prevents concurrent transactions that write the same data from both committing.
- · SSD DRAM cache
SSD DRAM is volatile working memory that can hold address mappings, controller metadata or buffered user data, depending on the design.
- · STATX_WRITE_ATOMIC
STATX_WRITE_ATOMIC is a Linux statx request for a file's reported atomic-write size limits and maximum vector count.
- · Storage cache flush
A storage cache flush requests persistence of writes buffered in the addressed storage stack, with its coverage and ordering defined by that interface.
- · Synchronous replication
Synchronous replication makes successful completion depend on acknowledgement that a specified set of replicas has reached the processing stage required by the replication contract.
- · Uncorrectable bit error rate
Uncorrectable bit error rate is a storage reliability metric that normalizes unrecovered data-error outcomes by a specified bit count after the stated correction and recovery procedures.
- · Value log
A value log is an append-oriented record store whose values are located through a separate index.
- · Write amplification
Device-level byte write amplification is the ratio of bytes programmed to the storage medium to bytes written by the host over the same accounting interval.
- · Write skew
Write skew is a concurrency anomaly in which transactions read overlapping data and make disjoint writes whose combined outcome can violate an invariant that each transaction would preserve alone.
- · Write-through cache
Write-through caching propagates writes to the backing store as part of completing them at the cache layer.
- · ALUA
Asymmetric Logical Unit Access is the SCSI mechanism for reporting logical-unit access states through target port groups.
- · Chunk fragmentation
Chunk fragmentation is the dispersion of chunks needed by a deduplicated restore across storage containers or an order poorly matched to the restore sequence.
- · Closed-loop load
A closed-loop load model uses a fixed client population whose next request follows completion of a previous request, optionally after think time.
- · Container capping
Container capping limits the old containers referenced by a backup segment by storing selected duplicate chunks again.
- · Coordinated omission
Coordinated omission is a measurement bias in which a workload generator or sampler misses observations because its activity depends on the slow responses being measured.
- · Disk defect skipping
Disk defect skipping formats a record around a known defective medium region so that normal reads bypass a gap instead of storing customer data there.
- · Fencing token
A fencing token identifies an ownership generation that a receiving service checks to reject operations from an obsolete owner.
- · Forward assembly area
A forward assembly area buffers requested chunks in future restore-output order using advance knowledge from the backup recipe.
- · Full-stripe write
A parity-array write supplying replacement data for every data position of a stripe so its new parity can be computed without reading the prior stripe contents.
- · Locally repairable code
A locally repairable code is an erasure code that provides recovery of specified missing fragments from a small subset of surviving fragments.
- · Lost write
A lost write is a write reported as completed whose intended update is absent at the required storage location.
- · Misdirected write
A misdirected write stores an update at the wrong address or on the wrong device.
- · Multipath I/O
Multipath I/O manages multiple paths to the same storage device for failover and, where supported, load distribution.
- · NAND copyback
NAND copyback relocates page data through an on-chip buffer to a destination page without transferring the payload to an external controller.
- · no_path_retry
no_path_retry is a DM Multipath setting that controls queueing behavior when every path to a mapped device is unavailable.
- · Noise-predictive maximum-likelihood detection
Noise-predictive maximum-likelihood (NPML) detection incorporates noise prediction or whitening into sequence-detector branch metrics for a partial-response recording channel.
- · Open-loop load
An open-loop load model schedules request arrivals independently of earlier request completions.
- · Parity pollution
Parity pollution is propagation of incorrect data into redundancy information through parity calculations.
- · PRML detection
Partial-response maximum-likelihood (PRML) detection recovers a recorded symbol sequence by equalizing readback to a chosen partial-response target and evaluating sequence likelihoods under a channel model.
- · Read-modify-write
A storage update that reads existing data needed to preserve unchanged information, combines it with replacement data, and writes the modified data and any required redundancy.
- · Reconstruct-write
A parity-array update method that computes new parity from incoming replacement data and unchanged data from the same stripe instead of updating old parity from old overwritten data.
- · Repair locality
Repair locality is the number of helper fragments needed to recover a missing fragment under a specified erasure-code repair procedure.
- · Repair-read amplification
Repair-read amplification is the ratio of bytes read during recovery to the explicitly stated quantity of lost data or lost encoded fragments being restored.
- · Winchester disk architecture
Winchester disk architecture combines low-mass, lightly loaded flying heads with lubricated disks in a protected head-disk assembly; it does not require non-removable media.
- · Zoned namespace
A zoned namespace is an NVMe namespace whose logical address space follows the ZNS command set's zone organization and write constraints.
Versioning and data documentation · Editorial and correction policy