Glossary

This glossary includes terms for the IBM System Storage™ and other Resiliency Family products.

This glossary includes selected terms and definitions from:

The American National Standard Dictionary for Information Systems, ANSI X3.172–1990, copyright 1990 by the American National Standards Institute (ANSI), 11 West 42nd Street, New York, New York 10036. Definitions derived from this book have the symbol (A) after the definition.
IBM Terminology, which is available online at the following Web site: http://www-306.ibm.com/ibm/terminology/index.html. Definitions derived from this source have the symbol (GC) after the definition.
The Information Technology Vocabulary developed by Subcommittee 1, Joint Technical Committee 1, of the International Organization for Standardization and the International Electrotechnical Commission (ISO/IEC JTC1/SC1). Definitions derived from this book have the symbol (I) after the definition. Definitions taken from draft international standards, committee drafts, and working papers that the ISO/IEC JTC1/SC1 is developing have the symbol (T) after the definition, indicating that final agreement has not been reached among the participating National Bodies of SC1.

This glossary uses the following cross-reference forms:

See

Refers the reader to one of two kinds of related information:

A term that is the expanded form of an abbreviation or acronym. This expanded form of the term contains the full definition.
A synonym or more preferred term

See also

Refers the reader to one or more related terms.

Contrast with

Refers the reader to a term that has an opposite or substantively different meaning.

Numerics

750: A model of the Enterprise Storage Server featuring a 2-way processor with limited physical storage capacity. This model can be updated to the model 800.
800: A model of the Enterprise Storage Server featuring a standard processor or an optional Turbo processor. The Model 800 supports RAID 5, RAID 10, and 15000 rpm drives. Model 800 supersedes Model F20.
1750: The machine type for the IBM System Storage DS6000 series. Models for the DS6000 include the 511 and EX1.
2105: The machine number for the IBM TotalStorage Enterprise Storage Server. Models of the Enterprise Storage Server are expressed as the number 2105 followed by "Model <xxx>", such as 2105 Model 800. The 2105 Model 100 is an Enterprise Storage Server expansion enclosure that is typically referred to simply as the Model 100.
2107: A hardware machine type for the IBM System Storage DS8000 series. Hardware models for the 2107 include base units and expansion units 92E and 9AE.
2244: A function authorization machine type for the IBM System Storage DS8000 series. The 2244 function authorization machine type corresponds with the 2107 hardware machine type and is used only for purposes of billing and authorizing the licensed functions on the 2107. Function authorization models for the 2244 are related to the type of licensed functions that you order. For example, Model RMC is for the remote mirror and copy function on a 2107 storage unit.
3390: The machine number of an IBM disk storage system. The Enterprise Storage Server, when interfaced to IBM zSeries hosts, is set up to appear as one or more 3390 devices, with a choice of 3390-2, 3390-3, or 3390-9 track formats.
3990: The machine number of an IBM control unit.
7133: The machine number of an IBM disk storage system. The Model D40 and 020 drawers of the 7133 can be installed in the 2105-100 expansion enclosure of the ESS.

A

access: 1) To obtain computing services or data.; 2) In computer security, a specific type of interaction between a subject and an object that results in flow of information from one to the other.
access-any mode: One of the two access modes that can be set for the storage unit during initial configuration. It enables all fibre-channel-attached host systems with no defined access profile to access all logical volumes on the storage unit. With a profile defined in DS Storage Manager for a particular host, that host has access only to volumes that are assigned to the WWPN for that host. See also pseudo host and worldwide port name.
ACK: See request for acknowledgment and acknowledgment.
agent: A program that automatically performs some service without user intervention or on a regular schedule. See also subagent.
alert: A message or log that a storage unit generates as the result of error event collection and analysis. An alert indicates that a service action is required.
allegiance: For zSeries, a relationship that is created between a device and one or more channel paths during the processing of certain conditions. See also implicit allegiance, contingent allegiance, and reserved allegiance.
allocated storage: The space that is allocated to volumes but not yet assigned. Contrast with assigned storage.
American National Standards Institute (ANSI): An organization of producers, consumers, and general interest groups that establishes the procedures by which accredited organizations create and maintain voluntary industry standards in the United States. (A)
anonymous: In the DS Storage Manager, the label on an icon that represents all connections that are using fibre-channel adapters between the storage unit and hosts but are not completely defined to the storage unit. See also anonymous host, pseudo host, and access-any mode.
anonymous host: Synonym for pseudo host. Contrast with anonymous and pseudo host.
ANSI: See American National Standards Institute.
APAR: See authorized program analysis report. (GC)
API: See application programming interface.
application programming interface: An interface that allows an application program that is written in a high-level language to use specific data or functions of the operating system or another program.
arbitrated loop: A fibre-channel topology that enables the interconnection of a set of nodes. See also point-to-point connection and switched fabric.
array: An ordered collection, or group, of physical devices (disk drive modules) that is used to define logical volumes or devices. In the storage unit, an array is a group of disks that the user designates to be managed by the RAID technique. See also redundant array of independent disks.
ASCII: (American National Standard Code for Information Interchange) The standard code, using a coded character set consisting of 7-bit coded characters (8 bits including parity check), that is used for information interchange among data processing systems, data communication systems, and associated equipment. The ASCII set consists of control characters and graphic characters. (A) Some organizations, including IBM, have used the parity bit to expand the basic code set.
assigned storage: The space that is allocated to a volume and that is assigned to a port.
authorized program analysis report (APAR): A request for correction of a defect in a current release of an IBM-supplied program. (GC)
availability: The degree to which a system or resource is capable of performing its normal function. See data availability.

B

bay: The physical space that is used for installing SCSI and fibre-channel host adapter cards. The DS6000™ has four bays, two in each cluster. See also service boundary.
bit: The smallest unit of computer information, which has two possible states that are represented by the binary digits 0 or 1. See also byte.
block: A string of data elements recorded or transmitted as a unit. The elements may be characters, words, or physical records. (GC); A group of consecutive bytes used as the basic storage unit in fixed-block architecture (FBA). All blocks on the storage device are the same size (fixed size). See also fixed-block architecture and data record.
byte: A string that represents a character and usually consists of eight binary digits that are treated as a unit. A byte is the smallest unit of storage that can be addressed directly. (GC) See also bit.

C

cache

A special-purpose buffer storage, smaller and faster than main storage, used to hold a copy of instructions and data obtained from main storage and likely to be needed next by the processor. (GC)

cache fast write

A form of the fast-write operation in which the storage server writes the data directly to cache, where it is available for later destaging.

cache hit

An event that occurs when a read operation is sent to the cluster, and the requested data is found in cache. Contrast with cache miss.

cache memory

Memory, typically volatile memory, that a storage server uses to improve access times to instructions or data. The cache memory is typically smaller and faster than the primary memory or storage medium. In addition to residing in cache memory, the same data also resides on the storage devices in the storage unit.

cache miss

An event that occurs when a read operation is sent to the cluster, but the data is not found in cache. Contrast with cache hit.

call home

A communication link established between the storage product and a service provider. The storage product can use this link to place a call to IBM or to another service provider when it requires service. With access to the machine, service personnel can perform service tasks, such as viewing error logs and problem logs or initiating trace and dump retrievals. (GC) See also heartbeat and remote technical assistance information network.

cascading

1) Connecting network controllers to each other in a succession of levels to concentrate many more lines than a single level permits.

2) In high-availability cluster multiprocessing (HACMP), pertaining to a cluster configuration in which the cluster node with the highest priority for a particular resource acquires the resource if the primary node fails. The cluster node relinquishes the resource to the primary node upon reintegration of the primary node into the cluster.

catcher

A server that service personnel use to collect and retain status data that an DS6000 sends to it.

CCR

See channel command retry.

CCW

See channel command word.

CD

See compact disc.

central electronics complex

The set of hardware facilities that are associated with a host computer.

channel

The part of a channel subsystem that manages a single I/O interface between a channel subsystem and a set of control units.

channel command retry (CCR)

The protocol used between a channel and a control unit that enables the control unit to request that the channel reissue the current command.

channel command word (CCW)

A data structure that specifies an I/O operation to the channel subsystem.

channel path

The interconnection between a channel and its associated control units.

channel subsystem

The part of a host computer that manages I/O communication between the program and any attached control units.

channel-subsystem image

For zSeries, the logical functions that a system requires to perform the function of a channel subsystem.

CKD

See count key data.

CLI

See command-line interface. See also Copy Services command-line interface.

cluster

1) A partition capable of performing all DS6000 functions. With two clusters in the DS6000, any operational cluster can take over the processing of a failing cluster.

cluster processor complex

The unit within a cluster that provides the management function for the DS6000. It consists of cluster processors, cluster memory, and related logic.

command-line interface (CLI)

An interface that defines a set of commands and enables a user (or a script-like language) to issue these commands by typing text in response to the command prompt (for example, DOS commands or UNIX shell commands). See also Copy Services command-line interface.

compact disc

An optically read disc, typically storing approximately 660 MB. CD-ROM (compact disc read-only memory) refers to the read-only format used to distribute DS6000 code and documentation.

compression

1) The process of eliminating gaps, empty fields, redundancies, and unnecessary data to shorten the length of records or blocks.

2) Any encoding that reduces the number of bits used to represent a given message or record. (GC)

concurrent copy

A facility on a storage server that enables a program to make a backup of a data set while the logical volume remains available for subsequent processing. The data in the backup copy is frozen at the point in time that the server responds to the request.

concurrent installation of licensed internal code

Process of installing licensed internal code on a DS6000 while applications continue to run.

concurrent maintenance

Service that is performed on a unit while it is operational.

concurrent media maintenance

Service performed on a disk drive module (DDM) without losing access to the data.

configure

In storage, to define the logical and physical devices, optional features, and program products of the input/output subsystem through the user interface that the storage unit provides for this function.

consistency group

A group of volumes participating in FlashCopy relationships in a logical subsystem, across logical subsystems, or across multiple storage units that must be kept in a consistent state to ensure data integrity.

consistency group interval time

The value in seconds that indicates the length of time between the formation of consistency groups.

consistent copy

A copy of a data entity (a logical volume, for example) that contains the contents of the entire data entity at a single instant in time.

console

A user interface to a server, for example, the interface provided on a personal computer. See also IBM^® System Storage™ Management Console.

contingent allegiance

In mainframe computing, a relationship that is created in a control unit between a device and a channel when the channel accepts unit-check status. The allegiance causes the control unit to guarantee access; the control unit does not present the busy status to the device. The allegiance enables the channel to retrieve sense data that is associated with the unit-check status on the channel path associated with the allegiance. (GC)

control path

The route that is established from the master storage unit to the subordinate storage unit when more than one storage unit participates in a Global Mirror session. If there is only one storage unit (the master) in the Global Mirror session, no control path is required.

control unit (CU)

1) A device that coordinates and controls the operation of one or more input/output devices, and synchronizes the operation of such devices with the operation of the system as a whole.

2) For zSeries, a storage server with ESCON or OEMI interfaces. The control unit adapts a native device interface to an I/O interface that a zSeries host system supports.

3) The portion of the storage unit that supports the attachment of emulated count key data devices over FICON or OEMI interfaces. See also cluster.

control-unit-initiated reconfiguration (CUIR)

A software mechanism that the DS6000 uses to request that an operating system of a zSeries host verify that one or more subsystem resources can be taken offline for service. The DS6000 can use this process to automatically vary channel paths offline and online to facilitate bay service or concurrent code installation. Depending on the operating system, support for this process might be model dependent, might depend on the IBM TotalStorage Enterprise Storage Server Subsystem Device Driver, or might not exist.

Coordinated Universal Time (UTC)

The international standard of time that is kept by atomic clocks around the world.

Copy Services

A collection of optional software features, with a Web-browser interface, used for configuring, managing, and monitoring data-copy functions.

Copy Services CLI

See Copy Services command-line interface.

Copy Services domain

See Copy Services server group.

Copy Services client

Software that runs on each DS6000 cluster in the Copy Services server group and that performs the following functions:

Communicates configuration, status and connectivity information to the Copy Services server
Performs data-copy functions on behalf of the Copy Services server

Copy Services command-line interface (Copy Services CLI)

The command-line interface software that is provided with DS6000 Copy Services and used for invoking Copy Services functions from host systems attached to the DS6000. See also command-line interface.

Copy Services server

A cluster that the Copy Services administrator designates to perform the DS6000 Copy Services functions.

Copy Services server group

A collection of user-designated DS6000 clusters participating in Copy Services functions that a designated, active, Copy Services server manages. A Copy Services server group is also called a Copy Services domain. See also domain.

count field

The first field of a count key data (CKD) record. This eight-byte field contains a four-byte track address (CCHH). It defines the cylinder and head that are associated with the track, and a one-byte record number (R) that identifies the record on the track. It defines a one-byte key length that specifies the length of the record's key field (0 means no key field). It defines a two-byte data length that specifies the length of the record's data field (0 means no data field). Only the end-of-file record has a data length of zero.

count key data (CKD)

In mainframe computing, a data-record format employing self-defining record formats in which each record is represented by up to three fields: a count field that identifies the record and specifies its format, an optional key field that identifies the data area contents, and an optional data field that typically contains the user data. For CKD records on the storage unit, the logical volume size is defined in terms of the device emulation mode (3390 or 3380 track format). The count field is always 8 bytes long and contains the lengths of the key and data fields, the key field has a length of 0 to 255 bytes, and the data field has a length of 0 to 65 535 or the maximum that will fit on the track. See also data record.

CPC

See cluster processor complex.

CRC

See cyclic redundancy check.

CU

See control unit.

CUIR

See control-unit initiated reconfiguration.

custom volume

A volume in count-key-data (CKD) format that is not a standard volume, which means that it does not necessarily present the same number of cylinders and capacity to its assigned logical control unit as provided by one of the following standard zSeries volume types: 3390-2, 3390-3, 3390-9, 3390-2 (3380-track mode), or 3390-3 (3380-track mode). See also count-key-data, interleave, standard volume, and volume.

CUT

See Coordinated Universal Time.

cyclic redundancy check (CRC)

A redundancy check in which the check key is generated by a cyclic algorithm. (T)

cylinder

A unit of storage on a CKD device with a fixed number of tracks.

D

DA: See device adapter.
daisy chain: See serial connection.
DASD: See direct access storage device.
DASD fast write (DFW): A function of a storage server in which active write data is stored in nonvolatile cache, thus avoiding exposure to data loss.
data availability: The degree to which data is available when needed, typically measured as a percentage of time that the system would be capable of responding to any data request (for example, 99.999% available).
data compression: A technique or algorithm used to encode data such that the encoded result can be stored in less space than the original data. The original data can be recovered from the encoded result through a reverse technique or reverse algorithm. See also compression.
Data Facility Storage Management Subsystem (DFSMS): An operating environment that helps automate and centralize the management of storage. To manage storage, DFSMS provides the storage administrator with control over data class, storage class, management class, storage group, and automatic class selection routine definitions.
data field: The optional third field of a count key data (CKD) record. The count field specifies the length of the data field. The data field contains data that the program writes.
data record: The basic unit of zSeries storage on a DS6000, also known as a count-key-data (CKD) record. Data records are stored on a track. The records are sequentially numbered starting with 0. The first record, R0, is typically called the track descriptor record and contains data that the operating system normally uses to manage the track. See also count-key-data and fixed-block architecture.
data set FlashCopy: For zSeries hosts, a feature of FlashCopy that indicates how many partial volume FlashCopy relationships are active on a volume.
data sharing: The ability of multiple host systems to concurrently utilize data that they store on one or more storage devices. The storage unit enables configured storage to be accessible to any, or all, attached host systems. To use this capability, the host program must be designed to support data that it is sharing.
DDM: See disk drive module.
DDM group: See disk pack.
dedicated storage: Storage within a storage unit that is configured such that a single host system has exclusive access to the storage.
demote: To remove a logical data unit from cache memory. A storage server demotes a data unit to make room for other logical data units in the cache or because the logical data unit is not valid. The storage unit must destage logical data units with active write units before they can be demoted. See also destage.
destage: To move data from an online or higher priority to an offline or lower priority device. The storage unit stages incoming data into cache and then destages it to disk.
device: For zSeries, a disk drive.
device adapter (DA): A physical component of the DS6000 that provides communication between the clusters and the storage devices. The DS6000 has eight device adapters that it deploys in pairs, one from each cluster. Device adapter pairing enables the DS6000 to access any disk drive from either of two paths, providing fault tolerance and enhanced availability.
device ID: The unique two-digit hexadecimal number that identifies the logical device.
device interface card: A physical subunit of a storage cluster that provides the communication with the attached device drive modules.
device number: For zSeries, a four-hexadecimal-character identifier, for example 13A0, that the systems administrator associates with a device to facilitate communication between the program and the host operator. The device number is associated with a subchannel.
device sparing: A subsystem function that automatically copies data from a failing device drive module to a spare device drive module. The subsystem maintains data access during the process.
DFS: See distributed file service.
DFSMS: See Data Facility Storage Management Subsystem.
direct access storage device (DASD): 1) A mass storage medium on which a computer stores data.; 2) A disk device.
disk cage: A container for disk drives. Each disk cage supports eight disk packs (64 disks).
disk drive: Standard term for a disk-based nonvolatile storage medium. The DS6000 uses hard disk drives as the primary nonvolatile storage media to store host data.
disk drive module (DDM): A field replaceable unit that consists of a single disk drive and its associated packaging.
disk drive module group: See disk pack.
disk drive set: A specific number of identical disk drives that have the same physical capacity and rpm.
disk pack: A group of disk drive modules (DDMs) installed as a unit in a DDM bay.
disk group: A collection of 4 disk drives that are connected to the same pair of IBM Serial Storage adapters and can be used to create a RAID array. A disk group can be formatted as count key data or fixed block, and as RAID or non-RAID, or it can be left unformatted. A disk group is a logical assemblage of disk drives. Contrast with disk pack.
distributed file service (DFS): A service that provides data access over IP networks.
DNS: See domain name system.
domain: 1) That part of a computer network in which the data processing resources are under common control.; 2) In TCP/IP, the naming system used in hierarchical networks.; 3) A Copy Services server group, in other words, the set of clusters the user designates to be managed by a particular Copy Services server.
domain name system (DNS): In TCP/IP, the server program that supplies name-to-address translation by mapping domain names to internet addresses. The address of a DNS server is the internet address of the server that hosts the DNS software for the network.
dotted decimal notation: A convention used to identify IP addresses. The notation consists of four 8-bit numbers written in base 10. For example, 9.113.76.250 is an IP address that contains the octets 9, 113, 76, and 250.
drawer: A unit that contains multiple device drive modules and provides power, cooling, and related interconnection logic to make the device drive modules accessible to attached host systems.
drive: 1) A peripheral device, especially one that has addressed storage media. See also disk drive module.; 2) The mechanism used to seek, read, and write information on a storage medium.
DS6000: See IBM System Storage DS6000.
DS6000 Batch Configuration tool: A program that automatically configures a DS6000. The configuration is based on data that IBM service personnel enter into the program.
DS Storage Manager: See IBM System Storage DS Storage Manager.
duplex: 1) Regarding Copy Services, the state of a volume pair after Remote Mirror and Copy has completed the copy operation and the volume pair is synchronized.; 2) In general, pertaining to a communication mode in which data can be sent and received at the same time.
dynamic sparing: The ability of a storage server to move data from a failing disk drive module (DDM) to a spare DDM while maintaining storage functions.

E

E10: The predecessor of the F10 model of the Enterprise Storage Server. See also F10.
E20: The predecessor of the F20 model of the Enterprise Storage Server. See also F20.
EC: See engineering change.
ECKD: See extended count key data.
eight pack: See disk pack.
electrostatic discharge (ESD): An undesirable discharge of static electricity that can damage equipment and degrade electrical circuitry.
emergency power off (EPO): A means of turning off power during an emergency, usually a switch.
enclosure: A unit that houses the components of a storage subsystem, such as a control unit, disk drives, and power source.
end of file: A coded character recorded on a data medium to indicate the end of the medium. On a count-key-data direct access storage device, the subsystem indicates the end of a file by including a record with a data length of zero.
engineering change (EC): An update to a machine, part, or program.
Enterprise Systems Architecture/390 (ESA/390): An IBM architecture for mainframe computers and peripherals. Processor systems that follow the ESA/390 architecture include the ES/9000® family. See also z/Architecture.
EPO: See emergency power off.
ERDS: See error-recording data set.
error-recording data set (ERDS): On zSeries hosts, a data set that records data-storage and data-retrieval errors. A service information message (SIM) provides the error information for the ERDS.
error recovery procedure: Procedures designed to help isolate and, where possible, to recover from errors in equipment. The procedures are often used in conjunction with programs that record information on machine malfunctions.
ESA/390: See Enterprise Systems Architecture/390.
ESD: See electrostatic discharge.
eServer: See IBM eServer.
ESSNet: See IBM TotalStorage Enterprise Storage Server Network.
express configuration: A method for configuring a storage complex, where the storage server simplifies the task by making several configuration decisions for the user.
extended count key data (ECKD): An extension of the count key data (CKD) architecture.
extent: A continuous space on a disk that is occupied by or reserved for a particular data set, data space, or file. The unit of increment is a track. See also multiple allegiance and parallel access volumes.
extent pool: A groups of extents. See also extent.

F

fabric: In fibre channel technology, a routing structure, such as a switch, receives addressed information and routes to the appropriate destination. A fabric can consist of more than one switch. When multiple fibre-channel switches are interconnected, they are said to be cascaded.
failback: Pertaining to a cluster recovery from failover following repair. See also failover.
failover: Pertaining to the process of transferring all control to a single cluster when the other cluster in the storage unit fails. See also cluster and failback.
fast write: A write operation at cache speed that does not require immediate transfer of data to a disk drive. The subsystem writes the data directly to cache, to nonvolatile storage, or to both. The data is then available for destaging. A fast-write operation reduces the time an application must wait for the I/O operation to complete.
FATA: See fibre-channel ATA.
FBA: See fixed-block architecture.
FC: See feature code. Note: FC is a common abbreviation for fibre channel in the industry, but the DS6000 customer documentation library reserves FC for feature code.
FC-AL: See Fibre Channel ATA.
FCP: See Fibre Channel Protocol.
FCS: See Fibre Channel standard.
feature code (FC): A code that identifies a particular orderable option and that is used by service personnel to process hardware and software orders. Individual optional features are each identified by a unique feature code.
fibre channel: A data-transmission architecture based on the ANSI Fibre Channel standard, which supports full-duplex communication. The DS6000 supports data transmission over fiber-optic cable through its fibre-channel adapters. See also Fibre Channel Protocol and Fibre Channel standard.
fibre-channel ATA (FATA): A hard drive that combines a fibre channel interface with an ATA drive. FATAs, which provide the high performance and capacity of an ATA drive, can be used wherever fibre channel drives can connect.
Fibre Channel Arbitrated Loop (FC-AL): An implementation of the Fibre Channel Standard that uses a ring topology for the communication fabric. Refer to American National Standards Institute (ANSI) X3T11/93-275. In this topology, two or more fibre-channel end points are interconnected through a looped interface. This topology directly connects the storage unit to an open systems host without going through a fabric switch.
Fibre Channel Connection (FICON): A fibre-channel communications protocol that is designed for IBM mainframe computers and peripherals. It connects the storage unit to one or more S/390 hosts using a FICON S/390 channel either directly or through a FICON switch.
Fibre Channel Protocol (FCP): A protocol used in fibre-channel communications with five layers that define how fibre-channel ports interact through their physical links to communicate with other ports.
Fibre Channel standard (FCS): An ANSI standard for a computer peripheral interface. The I/O interface defines a protocol for communication over a serial interface that configures attached units to a communication fabric. The protocol has two layers. The IP layer defines basic interconnection protocols. The upper layer supports one or more logical protocols (for example, FCP for SCSI command protocols and SBCON for zSeries command protocols). Refer to American National Standards Institute (ANSI) X3.230-199x. See also Fibre Channel Protocol.
fibre-channel topology: An interconnection topology supported on fibre-channel adapters. See also point-to-point connection, switched fabric, and arbitrated loop.
Fibre Channel Switched Fabric (FC-SF): An implementation of the Fibre Channel Standard that connects the storage unit to one or more open systems hosts through a fabric switch or connects one or more S/390 hosts that run LINUX on an Fibre Channel Protocol S/390 channel.
FICON: See fibre-channel connection.
FiconNet: In the DS Storage Manager, the label on a pseudo host icon that represents a host connection that uses the FICON protocol and that is not completely defined on the DS6000. See also pseudo host and access-any mode.
field replaceable unit (FRU): An assembly that is replaced in its entirety when any one of its components fails. In some cases, a field replaceable unit might contain other field replaceable units. (GC)
FIFO: See first-in-first-out.
File Transfer Protocol (FTP): In TCP/IP, an application protocol used to transfer files to and from host computers. See also Transmission Control Protocol/Internet Protocol.
firewall: A protection against unauthorized connection to a computer or a data storage system. The protection is usually in the form of software on a gateway server that grants access to users who meet authorization criteria.
first-in-first-out (FIFO): A queuing technique in which the next item to be retrieved is the item that has been in the queue for the longest time. (A)
fixed-block architecture (FBA): An architecture for logical devices that specifies the format of and access mechanisms for the logical data units on the device. The logical data unit is a block. All blocks on the device are the same size (fixed size). The subsystem can access them independently.
fixed-block device: An architecture for logical devices that specifies the format of the logical data units on the device. The logical data unit is a block. All blocks on the device are the same size (fixed size); the subsystem can access them independently. This is the required format of the logical data units for host systems that attach with a SCSI or fibre-channel interface. See also fibre channel and small computer systems interface.
FlashCopy: An optional feature of the DS6000 that can make an instant copy of data, that is, a point-in-time copy of a volume.
FlashCopy relationship: A mapping of a FlashCopy source volume and a FlashCopy target volume that allows a point-in-time copy of the source volume to be copied to the target volume. FlashCopy relationships exist from the time that you initiate a FlashCopy operation until the storage unit copies all data from the source volume to the target volume or until you delete the FlashCopy relationship, if it is persistent.
FRU: See field replaceable unit.
FTP: See File Transfer Protocol.
full duplex: See duplex.
fuzzy copy: A function of the Global Copy feature wherein modifications to the primary logical volume are performed on the secondary logical volume at a later time. The original order of update is not strictly maintained. See also Global Copy.

G

GB: See gigabyte.
GDPS: See Geographically Dispersed Parallel Sysplex.
Geographically Dispersed Parallel Sysplex (GDPS): A zSeries multisite application-availability solution.
gigabyte (GB): A gigabyte of storage is 10⁹ bytes. A gigabyte of memory is 2³⁰ bytes.
Global Copy: An optional capability of the DS6000 remote mirror and copy feature that maintains a fuzzy copy of a logical volume on the same DS6000 or on another DS6000. In other words, all modifications that any attached host performs on the primary logical volume are also performed on the secondary logical volume at a later point in time. The original order of update is not strictly maintained. See also Remote Mirror and Copy and Metro Mirror.
Global Mirror: An optional capability of the remote mirror and copy feature that provides a 2-site extended distance remote copy. Data that is written by the host to the storage unit at the local site is automatically maintained at the remote site. See also Metro Mirror and Remote Mirror and Copy.
group: In DS6000 documentation, a nickname for two different kinds of groups, depending on the context. See disk pack or Copy Services server group.

H

HA: See host adapter.
HACMP: See high availability cluster multiprocessing.
hard disk drive (HDD): 1) A storage medium within a storage server used to maintain information that the storage server requires.; 2) A mass storage medium for computers that is typically available as a fixed disk (such as the disks used in system units of personal computers or in drives that are external to a personal computer) or a removable cartridge.
hardware service manager: An option on an AS/400 or iSeries host that enables the user to display and work with system hardware resources and to debug input-output processors (IOP), input-output adapters (IOA), and devices.
HCD: See Hardware Configuration Data.
HDA: See head disk assembly.
HDD: See hard disk drive.
hdisk: An AIX term for storage space.
head disk assembly (HDA): The portion of an HDD associated with the medium and the read/write head.
heartbeat: A status report sent at regular intervals from the DS6000. The service provider uses this report to monitor the health of the call home process. See also call home, heartbeat call home record, and remote technical assistance information network.
heartbeat call home record: Machine operating and service information sent to a service machine. These records might include such information as feature code information and product logical configuration information.
hierarchical storage management: 1) A function in storage management software, such as Tivoli Storage Management or Data Facility Storage Management Subsystem/MVS (DFSMS/MVS), that automatically manages free space based on the policy that the storage administrator sets.; 2) In AS/400 storage management, an automatic method to manage and distribute data between the different storage layers, such as disk units and tape library devices.
high availability cluster multiprocessing (HACMP): Software that provides host clustering, so that a failure of one host is recovered by moving jobs to other hosts within the cluster.
high-speed loop (HSL): A hardware connectivity architecture that links system processors to system input/output buses and other system units.
home address: A nine-byte field at the beginning of a track that contains information that identifies the physical track and its association with a cylinder.
hop: Interswitch connection. A hop count is the number of connections that a particular block of data traverses between source and destination. For example, data traveling from one hub over a wire to another hub traverses one hop.
host: See host system.
host adapter: A physical subunit of a storage server that provides the ability to attach to one or more host I/O interfaces.
host name: The Internet address of a machine in the network. The host name can be entered in the host definition as the fully qualified domain name of the attached host system, such as mycomputer.city.company.com, or as the subname of the fully qualified domain name, for example, mycomputer. See also host system.
host processor: A processor that controls all or part of a user application network. In a network, the processing unit in which the data communication access method resides. See also host system.
host system: A computer, either of the mainframe (for example, zSeries) or of the open-systems type, that is connected to the DS6000. Hosts are connected through FICON or fibre-channel interfaces.
hot plug: Pertaining to the ability to add or remove a hardware facility or resource to a unit while power is on.
HSL: See high-speed loop.

I

i5/OS: The IBM operating system that runs the IBM i5/OS and eServer i5 server families of servers.
IBM eServer: The IBM brand name for a series of server products that are optimized for e-commerce. The products include the iSeries, pSeries, xSeries, and zSeries.
IBM product engineering (PE): The third-level of IBM service support. Product engineering is composed of IBM engineers who have experience in supporting a product or who are knowledgeable about the product.
IBM Serial Storage adapter: A physical adapter based on the IBM Serial Storage architecture. IBM Serial Storage adapters connect disk drive modules to DS6000 clusters.
IBM System Storage: The brand name used to identify storage products from IBM, including the IBM System Storage DS6000. See also IBM System Storage DS6000 and IBM System Storage DS Storage Manager.
IBM System Storage DS6000: A member of the IBM System Storage Resiliency Family of storage servers and attached storage devices (disk drive modules). The DS6000 delivers high-performance, fault-tolerant storage and management of enterprise data, affording access through multiple concurrent operating systems and communication protocols. High performance is provided by multiple symmetrical multiprocessors, integrated caching, RAID support for the disk drive modules, and disk access through a high-speed serial storage architecture interface.
IBM System Storage DS CLI: The command-line interface (CLI) that is specific to the DS6000.
IBM System Storage DS Storage Manager (DS Storage Manager): Software with a Web-browser interface for configuring the DS6000.
IBM HyperPAV: See HyperPAV.
IBM TotalStorage Enterprise Storage Server Network (ESSNet): A private network providing Web browser access to the Enterprise Storage Server. IBM installs the ESSNet software on an IBM workstation called the IBM TotalStorage ESS Master Console, supplied with the first ESS delivery.
IBM System Storage Management Console (MC): An IBM workstation that acts as the focal point for configuration, Copy Services management, and maintenance for the DS6000 . It includes a Web browser that provides links to the user interface, including the DS Storage Manager and the DS6000 Copy Services.
IBM System Storage Multipath Subsystem Device Driver (SDD): IBM software that provides multipath configuration support for a host system that is attached to storage devices. SDD provides enhanced data availability, dynamic input/output load balancing across multiple paths, and automatic path failover protection.
IBM System Storage Resiliency Family: A set of hardware and software features and products, as well as integrated software and services that are available on the IBM System Storage DS6000 and the IBM TotalStorage Enterprise Storage Server, Models 750 and 800.
IML: See initial microcode load.
implicit allegiance: In Enterprise Systems Architecture/390, a relationship that a control unit creates between a device and a channel path when the device accepts a read or write operation. The control unit guarantees access to the channel program over the set of channel paths that it associates with the allegiance.
initial microcode load (IML): The action of loading microcode for a computer into that computer's storage.
initial program load (IPL): The action of loading software into a computer, typically an operating system that controls the computer.
initiator: A SCSI device that communicates with and controls one or more targets. Contrast with target.
i-node: The internal structure in an AIX operating system that describes the individual files in the operating system. It contains the code, type, location, and owner of a file.
input/output (I/O): Pertaining to (a) input, output, or both or (b) a device, process, or channel involved in data input, data output, or both.
input/output configuration data set: A configuration definition built by the I/O configuration program (IOCP) and stored on disk files associated with the processor controller.
interleave: To automatically create two striped partitions across the drives in a RAID-5 array, both of which use the count-key-data (CKD) record format.
Internet Protocol (IP): In the Internet suite of protocols, a protocol without connections that routes data through a network or interconnecting networks and acts as an intermediary between the higher protocol layers and the physical network. The upper layer supports one or more logical protocols (for example, a SCSI-command protocol and a zSeries command protocol). Refer to ANSI X3.230-199x. The IP acronym is the IP in TCP/IP. See also Transmission Control Protocol/Internet Protocol.
invalidate: To remove a logical data unit from cache memory because it cannot support continued access to the logical data unit on the device. This removal might be the result of a failure within the storage server or a storage device that is associated with the device.
I/O: See input/output.
I/O adapter (IOA): An input-output adapter on the PCI bus.
IOCDS: See input/output configuration data set.
IOCP: See I/O Configuration Program.
I/O Configuration Program (IOCP): A program that defines to a system all the available I/O devices and channel paths.
I/O device: An addressable read and write unit, such as a disk drive device, magnetic tape device, or printer.
I/O interface: An interface that enables a host to perform read and write operations with its associated peripheral devices.
I/O Priority Queueing: A facility in the Workload Manager of zSeries that enables the system administrator to set priorities for queueing I/Os from different system images. See also multiple allegiance and parallel access volumes.
I/O processor (IOP): Controls input-output adapters and other devices.
I/O sequential response time: The time an I/O request is queued in processor memory waiting for previous I/Os to the same volume to complete.
IP: See Internet Protocol.
IPL: See initial program load.
iSeries: An IBM eServer product that emphasizes integration. It is the successor to the AS/400 family of servers.

J

Java Virtual Machine (JVM): A software implementation of a central processing unit (CPU) that runs compiled Java code (applets and applications). (GC)
JVM: See Java Virtual Machine.

K

KB: See kilobyte.
key field: The second (optional) field of a count key data record. The key length is specified in the count field. The key length determines the field length. The program writes the data in the key field and uses the key field to identify or locate a given record. The subsystem does not use the key field.
kilobyte (KB): 1) For processor storage, real, and virtual storage, and channel volume, 2¹⁰ or 1024 bytes.; 2) For disk storage capacity and communications volume, 1000 bytes.
Korn shell: Interactive command interpreter and a command programming language.
KPOH: See thousands of power-on hours.

L

LAN: See local area network.
last-in first-out (LIFO): A queuing technique in which the next item to be retrieved is the item most recently placed in the queue. (A)
LBA: See logical block address.
LCU: See logical control unit.
least recently used (LRU): 1) The algorithm used to identify and make available the cache space that contains the least-recently used data.; 2) A policy for a caching algorithm that chooses to remove from cache the item that has the longest elapsed time since its last access.
LED: See light-emitting diode.
licensed machine code: Microcode that IBM does not sell as part of a machine, but licenses to the customer. LMC is implemented in a part of storage that is not addressable by user programs. Some IBM products use it to implement functions as an alternate to hard-wired circuitry.
LIFO: See last-in first-out.
light-emitting diode (LED): A semiconductor chip that gives off visible or infrared light when activated.
local area network (LAN): A computer network located on a user's premises within a limited geographic area.
local e-mail: An e-mail configuration option for storage servers that are connected to a host-system network that does not have a domain name system (DNS) server.
logical address: On an ESCON interface, the portion of a source or destination address in a frame used to select a specific channel-subsystem or control-unit image.
logical block address (LBA): The address assigned by the DS6000 to a sector of a disk.
logical control unit (LCU): See control-unit image.
logical data unit: A unit of storage that is accessible on a given device.
logical device: The facilities of a storage server (such as the DS6000) associated with the processing of I/O operations directed to a single host-accessible emulated I/O device. The associated storage is referred to as a logical volume. The logical device is mapped to one or more host-addressable units, such as a device on a zSeries I/O interface or a logical unit on a SCSI I/O interface, such that the host initiating I/O operations to the I/O-addressable unit interacts with the storage on the associated logical device.
logical partition (LPAR): For zSeries, a set of functions that create the programming environment in which more than one logical partition (LPAR) is established on a processor. An LPAR is conceptually similar to a virtual machine environment except that the LPAR is a function of the processor. Also, the LPAR does not depend on an operating system to create the virtual machine environment. (DS8000 series only)
logical path: 1) The relationship between a channel image and a control-unit image that designates the physical path to be used for device-level communications between these images. The logical path is established as part of the channel and control-unit initialization procedures by the exchange of link-level frames.; 2) With the Remote Mirror and Copy feature, the relationship between a source logical subsystem (LSS) and a target LSS that is created over a physical path through the interconnection fabric that is used for Remote Mirror and Copy functions. An LSS is a primary control unit, which performs the functions of a channel image.
logical subsystem (LSS): A topological construct that consists of a group of up to 256 logical devices. A DS6000 can have (if CDK only) up to 32 CKD-formatted logical subsystems (8192 CKD logical devices) or (if FBA only) up to 32 fixed-block logical subsystems (8192 fixed-block logical devices). If mixed CKD and FBA, a DS6000 can have up to 16 CKD-formatted logical subsystems (4096 CKD logical devices) and up to 16 fixed-block logical subsystems (4096 fixed-block logical devices). The logical subsystem facilitates configuration of the DS6000 and might have other implications relative to the operation of certain functions. There is a one-to-one mapping between a CKD logical subsystem and a zSeries control-unit image.
For zSeries hosts, a logical subsystem represents a logical control unit (LCU). Each control-unit image is associated with only one logical subsystem. See also control-unit image.
logical unit: In open systems, a logical disk drive.
logical unit number (LUN): In the SCSI protocol, a unique number that is used on a SCSI bus to enable it to differentiate between separate devices, each of which is a logical unit.
logical volume: The storage medium that is associated with a logical disk drive. A logical volume typically resides on one or more storage devices. The DS6000 administrator defines this unit of storage. The logical volume, when residing on a RAID-formatted array, is spread over the drives in the array.
logical volume manager (LVM): A set of system commands, library routines, and other tools that allow the user to establish and control logical volume storage. The LVM maps data between the logical view of storage space and the physical disk drive module.
longitudinal redundancy check (LRC): 1) A method of error checking during data transfer that involves checking parity on a row of binary digits that are members of a set that forms a matrix. Longitudinal redundancy check is also called a longitudinal parity check.; 2) A mechanism that the DS6000 uses for locating errors. The LRC checks the data as it progresses from the host, through theDS6000 controller, into the device adapter, and to the array.
longwave laser adapter: A connector that is used between a host and the DS6000 to support longwave fibre-channel communication.
loop: The physical connection between a pair of device adapters in the DS6000. See also device adapter.
LPAR: See logical partition.
LRC: See longitudinal redundancy check.
LRU: See least recently used.
LSS: See logical subsystem.
LUN: See logical unit number.
LVM: See logical volume manager.

M

machine level control (MLC): A database that contains the EC level and configuration of products in the field.
machine reported product data (MRPD): Product data gathered by a machine and sent to a destination such as an IBM support server or RETAIN. These records might include such information as feature code information and product logical configuration information.
mainframe: A computer, usually in a computer center, with extensive capabilities and resources to which other computers may be connected so that they can share facilities. (T)
maintenance analysis procedure (MAP): A hardware maintenance document that gives an IBM service representative a step-by-step procedure for tracing a symptom to the cause of a failure.
management console: See IBM System Storage Management Console.
management information base (MIB): 1) A collection of objects that can be accessed by means of a network management protocol. (GC); 2) The MIB record conforms to the Open Systems Interconnection (OSI) standard defined by the International Organization for Standardization (ISO) for the exchange of information. See also simple network management protocol.
MAP: See maintenance analysis procedure.
master storage unit: The physical unit that controls the creation of consistency groups in a Global Mirror session. The master storage unit sends commands to subordinate storage units. A storage unit can be a master for only one Global Mirror session. Contrast with subordinate storage unit.
maximum consistency group drain time: The value in seconds that indicates the maximum time that writes from the local site are delayed to the remote site while the current consistency group is being formed at the remote site. When this time is exceeded, the current attempt to form a consistency group is ended and another attempt is started. If this time is exceeded five times, this maximum time is ignored on the next attempt to form a consistency group. The default value is the larger of four minutes or two times the consistency group interval time if this value is set to zero.
maximum coordination time: The value in milliseconds that indicates the maximum time that is allowed for host I/O to be delayed during the coordination of the primary volumes of an Global Mirror session. The default is 50 milliseconds if this value is set to zero.
MB: See megabyte.
MC: See IBM System Storage Management Console.
MCA: See Micro Channel architecture.
MDM: See Multiple Device Manager.
mean time between failures (MTBF): 1) A projection of the time that an individual unit remains functional. The time is based on averaging the performance, or projected performance, of a population of statistically independent units. The units operate under a set of conditions or assumptions.; 2) For a stated period in the life of a functional unit, the mean value of the lengths of time between consecutive failures under stated conditions. (I) (A)
medium: For a storage unit, the disk surface on which data is stored.
megabyte (MB): 1) For processor storage, real and virtual storage, and channel volume, 2²⁰ or 1 048 576 bytes.; 2) For disk storage capacity and communications volume, 1 000 000 bytes.
Metro Mirror: A function of a storage server that maintains a consistent copy of a logical volume on the same storage server or on another storage server. All modifications that any attached host performs on the primary logical volume are also performed on the secondary logical volume. See also Remote Mirror and Copy and Global Copy.
MES: See miscellaneous equipment specification.
MIB: See management information base.
Micro Channel architecture (MCA): The rules that define how subsystems and adapters use the Micro Channel bus in a computer. The architecture defines the services that each subsystem can or must provide.
Microsoft Internet Explorer: Web browser software manufactured by Microsoft.
migration: The replacement of a system or subsystem with a different type of system or subsystem, such as replacing a SCSI host adapter with a fibre-channel host adapter. In the context of data migration regarding the DS6000, the transfer of data from one storage unit to another, such as from a 3390 to the DS6000.
MIH: See missing-interrupt handler.
mirrored pair: Two units that contain the same data. The system refers to them as one entity.
mirroring: In host systems, the process of writing the same data to two disk units within the same auxiliary storage pool at the same time.
miscellaneous equipment specification (MES): IBM field-installed change to a machine.
missing-interrupt handler (MIH): An MVS and MVS/XA facility that tracks I/O interrupts. MIH informs the operator and creates a record whenever an expected interrupt fails to occur before a specified elapsed time is exceeded.
MLC: See machine level control.
mobile solutions terminal (MoST): The mobile terminal used by service personnel.
mode conditioning patch cable: A cable that converts a single-mode signal from a longwave adapter into a light signal that is appropriate for multimode fibre. Another mode conditioning patch cable is required at the terminating end of the multimode fibre to convert the signal back to a single-mode signal for a longwave adapter.
Model 100: A 2105 Model 100, often simply referred to as a Mod 100, is an expansion enclosure for the Enterprise Storage Server. See also 2105.
MoST: See mobile solutions terminal.
MRPD: See machine reported product data.
MSA: See multiport serial adapter.
MTBF: See mean time between failures.
Multipath Subsystem Device Driver: See IBM System Storage DS6000 Multipath Subsystem Device Driver.
multiple allegiance: A DS6000 hardware function that is independent of software support. This function enables multiple system images to concurrently access the same logical volume on the DS6000 as long as the system images are accessing different extents. See also extent and parallel access volumes.
Multiple Device Manager (MDM): A component of the IBM TotalStorage Productivity Center that allows administrators to configure, manage, and monitor the performance of SAN storage devices from a single console.
multiple relationship FlashCopy: An option of the DS6000 that creates backup copies from one source to multiple targets by simultaneously establishing multiple FlashCopy relationships.
multiple virtual storage (MVS): Implies MVS/390, MVS/XA, MVS/ESA, and the MVS element of the zSeries operating system.
multiplex: The action of transmitting simultaneously.
multiport serial adapter (MSA): An adapter on the IBM System Storage Management Console that has multiple ports to which aDS6000 can be attached.
multiprocessor: A computer that includes two or more processors that have common access to a main storage. For the DS6000, the multiprocessors operate in parallel.
MVS: See multiple virtual storage.

N

name server: A server that stores names of the participating DS6000 clusters.
near-line: A type of intermediate storage between online storage (which provides constant, rapid access to data) and offline storage (which provides infrequent data access for backup purposes or long-term storage).
Netfinity: IBM Intel-processor-based server; predecessor to the IBM xSeries server.
Netscape Navigator: Web browser software manufactured by Netscape.
network manager: A program or group of programs that is used to monitor, manage, and diagnose the problems of a network. (GC)
node: The unit that is connected in a fibre-channel network. A DS6000 is a node in a fibre-channel network.
non-RAID: A disk drive set up independently of other disk drives and not set up as part of a disk pack to store data using the redundant array of disks (RAID) data-striping methodology.
nonremovable medium: A recording medium that cannot be added to or removed from a storage device.
nonvolatile storage (NVS): Memory that stores active write data to avoid data loss in the event of a power loss.
NVS: See nonvolatile storage.

O

octet: In Internet Protocol addressing, one of the four parts of a 32-bit integer presented in dotted decimal notation. See also dotted decimal notation.
OEMI: See original equipment manufacturer's information.
open system: A system whose characteristics comply with standards made available throughout the industry and that therefore can be connected to other systems complying with the same standards. Applied to the DS6000, such systems are those hosts that connect to the DS6000 through SCSI or FCP protocols. See also small computer system interface and Fibre Channel Protocol.
organizationally unique identifier (OUI): An IEEE-standards number that identifies an organization with a 24-bit globally unique assigned number referenced by various standards. OUI is used in the family of 802 LAN standards, such as Ethernet and Token Ring.
original equipment manufacturer's information (OEMI): A reference to an IBM guideline for a computer peripheral interface. The interface uses ESA/390 logical protocols over an I/O interface that configures attached units in a multidrop bus topology.
OS/390: The IBM operating system that includes and integrates functions that many IBM software products (including the MVS operating system) previously provided for the IBM S/390 family of enterprise servers.
OUI: See organizationally unique identifier.

P

panel: The formatted display of information that appears on a display screen.
parallel access volumes (PAV): A licensed function of the DS6000 that enables OS/390 and z/OS systems to issue concurrent I/O requests against a count key data logical volume by associating multiple devices of a single control-unit image with a single logical device. Up to eight device addresses can be assigned to a PAV. The PAV function enables two or more concurrent write operations to the same logical volume, as long as the write operations are not to the same extents. See also extent, I/O Priority Queueing, and multiple allegiance.
parity: A data checking scheme used in a computer system to ensure the integrity of the data. The RAID implementation uses parity to re-create data if a disk drive fails.
path group: In zSeries architecture, a set of channel paths that are defined to a control unit as being associated with a single logical partition (LPAR). The channel paths are in a group state and are online to the host. See also logical partition.
path group identifier: In zSeries architecture, the identifier that uniquely identifies a given logical partition (LPAR). The path group identifier is used in communication between the LPAR program and a device. The identifier associates the path group with one or more channel paths, thereby defining these paths to the control unit as being associated with the same LPAR. See also logical partition.
PAV: See parallel access volumes.
PCI: See peripheral component interconnect.
PDU: See protocol data unit.
PE: See IBM product engineering.
peripheral component interconnect (PCI): An architecture for a system bus and associated protocols that supports attachments of adapter cards to a system backplane.
persistent FlashCopy: A state where a FlashCopy relationship remains indefinitely until the user deletes it. The relationship between the source and target volumes is maintained after a background copy completes.
physical path: A single path through the I/O interconnection fabric that attaches two units. For Copy Services, this is the path from a host adapter on one DS6000 (through cabling and switches) to a host adapter on anotherDS6000.
pinned data: Data that is held in cache until either an error condition is corrected and it can be moved to disk storage or until the data is discarded by a host command. Pinned data conditions can only occur on an ESS Model 800 during fast-write or dual-copy functions.
planar: The main printed circuit board (PCB) that other PCBs or assemblies plug into. The planar distributes both power and signals and therefore creates a common communications path to whichever device that plugs into it.
point-in-time copy: A FlashCopy option that creates an instantaneous view of original source data at a specific moment in time.
point-to-point connection: A fibre-channel topology that enables the direct interconnection of ports. See also arbitrated loop and switched fabric.
port: A physical connection on a host adapter to the cable that connects the DS6000 to hosts, switches, or another DS6000. The DS6000 uses SCSI and ESCON host adapters that have two ports per adapter, and fibre-channel host adapters that have one port. See also ESCON, fibre-channel, host adapter, and small computer system interface.
POST: See power-on self test.
power-on self test (POST): A diagnostic test that servers or computers run when they are turned on.
predictable write: A write operation that can cache without knowledge of the existing format on the medium. All write operations on FBA DASD devices are predictable. On CKD DASD devices, a write operation is predictable if it does a format write operation for the first data record on the track.
primary control unit: The DS6000 to which a Remote Mirror and Copy primary device is physically attached.
processor complex: A partition of a storage server that is capable of performing all defined functions of the storage server. Multiple processor complexes provide redundancy.
product engineering: See IBM product engineering.
program: On a computer, a generic term for software that controls the operation of the computer. Typically, the program is a logical assemblage of software modules that perform multiple related tasks.
program-controlled interruption: An interruption that occurs when an I/O channel fetches a channel command word with the program-controlled interruption flag on.
program temporary fix (PTF): A temporary solution to, or bypass of, a problem diagnosed by IBM as the result of a defect in a current unaltered release of a licensed program. (GC)
promote: To add a logical data unit to cache memory.
protected volume: In AS/400, a disk storage device that is protected from data loss by RAID techniques. An AS/400 host does not mirror a volume configured as a protected volume, while it does mirror all volumes configured as unprotected volumes. The DS6000, however, can be configured to indicate that an AS/400 volume is protected or unprotected and give it RAID protection in either case.
protocol data unit (PDU): A unit of data specified in the protocol of a given layer and consisting of protocol control information for the layer and, possibly, user data for the layer.
pSeries: The product name of an IBM eServer product that emphasizes performance. It is the successor to the RS/6000 family of servers.
pseudo host: A host connection that is not explicitly defined to the DS6000 and that has access to at least one volume that is configured on the DS6000. The FiconNet pseudo host icon represents the FICON protocol. The pseudo host icon labeled Anonymous represents hosts connected through the FCP protocol. Anonymous host is a commonly used synonym for pseudo host. The DS6000 adds a pseudo host icon only when it is set to access-any mode. See also access-any mode.
PTF: See program temporary fix.
PV Links: Short for Physical Volume Links, an alternate pathing solution from Hewlett-Packard that provides for multiple paths to a volume, as well as static load balancing.

R

R0: See track-descriptor record.
rack: See enclosure.
RAID: See redundant array of independent disks. RAID is also commonly expanded to redundant array of inexpensive disks. See also array.
RAID 5: A type of RAID that optimizes cost-effective performance while emphasizing use of available capacity through data striping. RAID 5 provides fault tolerance for up to two failed disk drives by distributing parity across all the drives in the array plus one parity disk drive. The DS6000 automatically reserves spare disk drives when it assigns arrays to a device adapter pair (DA pair). See also device adapter, RAID 10, and redundant array of independent disks.
RAID 10: A type of RAID that optimizes high performance while maintaining fault tolerance for up to two failed disk drives by striping volume data across several disk drives and mirroring the first set of disk drives on an identical set. The DS6000 automatically reserves spare disk drives when it assigns arrays to a device adapter pair (DA pair). See also device adapter, RAID 5, and redundant array of independent disks.
random access: A mode of accessing data on a medium in a manner that requires the storage device to access nonconsecutive storage locations on the medium.
rank: One or more arrays that are combined to create a logically contiguous storage space.
redundant array of independent disks (RAID): A methodology of grouping disk drives for managing disk storage to insulate data from a failing disk drive.
refresh FlashCopy target volume: An option (previously called incremental FlashCopy) of the DS6000 that creates a point-in-time data copy without copying an entire volume for each point-in-time copy.
Remote Mirror and Copy: A feature of a storage server that constantly updates a secondary copy of a logical volume to match changes made to a primary logical volume. The primary and secondary volumes can be on the same storage server or on separate storage servers. See also Global Mirror, Metro Mirror and Global Copy.
remote technical assistance information network (RETAIN): The initial service tracking system for IBM service support, which captures heartbeat and call-home records. See also support catcher and support catcher telephone number.
REQ/ACK: See request for acknowledgment and acknowledgment.
request for acknowledgment and acknowledgment (REQ/ACK): A cycle of communication between two data transport devices for the purpose of verifying the connection, which starts with a request for acknowledgment from one of the devices and ends with an acknowledgment from the second device. The REQ and ACK signals help to provide uniform timing to support synchronous data transfer between an initiator and a target. The objective of a synchronous data transfer method is to minimize the effect of device and cable delays.
reserved allegiance: For zSeries, a relationship that is created in a control unit between a device and a channel path, or path group, when the device completes a Sense Reserve command. The allegiance causes the control unit to guarantee access (that is, busy status is not presented) to the device. Access is over the set of channel paths that are associated with the allegiance; access is for one or more channel programs until the allegiance ends.
RETAIN: See remote technical assistance information network.

S

S/390: IBM enterprise servers based on Enterprise Systems Architecture/390 (ESA/390). S/390 is the currently accepted shortened form of the original name System/390.
S/390 storage: Storage arrays and logical volumes that are defined as connected to S/390 servers. This term is synonymous with count-key-data storage.
SAID: See system adapter identification number.
SAM: See sequential access method.
SAN: See storage area network.
SBCON: See Single-Byte Command Code Sets Connection.
screen: The physical surface of a display device upon which information is shown to users.
SCSI: See small computer system interface.
SCSI device: A disk drive connected to a host through an I/O interface using the SCSI protocol. A SCSI device is either an initiator or a target. See also initiator and small computer system interface.
SCSI-FCP: Synonym for Fibre Channel Protocol, a protocol used to transport data between an open-systems host and a fibre-channel adapter on an DS6000. See also Fibre Channel Protocol and small computer system interface.
SCSI host systems: Host systems that are attached to the DS6000 with a SCSI interface. Such host systems run on UNIX, i5/OS, Windows NT, Windows 2000, or Novell NetWare operating systems.
SCSI ID: A unique identifier assigned to a SCSI device that is used in protocols on the SCSI interface to identify or select the device. The number of data bits on the SCSI bus determines the number of available SCSI IDs. A wide interface has 16 bits, with 16 possible IDs.
SDD: See IBM Subsystem Multipathing Device Driver.
secondary control unit: The DS6000 to which a Remote Mirror and Copy secondary device is physically attached.
self-timed interface (STI): An interface that has one or more conductors that transmit information serially between two interconnected units without requiring any clock signals to recover the data. The interface performs clock recovery independently on each serial data stream and uses information in the data stream to determine character boundaries and inter-conductor synchronization.
sequential access: A mode of accessing data on a medium in a manner that requires the storage device to access consecutive storage locations on the medium.
sequential access method (SAM): An access method for storing, deleting, or retrieving data in a continuous sequence based on the logical order of the records in the file.
serial connection: A method of device interconnection for determining interrupt priority by connecting the interrupt sources serially.
server: A host that provides certain services to other hosts that are referred to as clients.; A functional unit that provides services to one or more clients over a network. (GC)
service boundary: A category that identifies a group of components that are unavailable for use when one of the components of the group is being serviced. Service boundaries are provided on the DS6000, for example, in each host bay and in each cluster.
service clearance: The area that is required to open the service covers and to pull out components for servicing.
service information message (SIM): A message sent by a storage server to service personnel through an zSeries operating system.
service personnel: A generalization referring to individuals or companies authorized to service the DS6000. The terms service provider, service representative, and IBM service support representative (SSR) refer to types of service personnel. See also service support representative.
service processor: A dedicated processing unit that is used to service a storage unit.
service support representative (SSR): Individuals or a company authorized to service the DS6000. This term also refers to a service provider, a service representative, or an IBM service support representative (SSR). An IBM SSR installs the DS6000.
SES: SCSI Enclosure Services.
session: A collection of volumes within a logical subsystem that are managed together during the creation of consistent copies of data. All volumes in a session must transfer their data successfully to the remote site before the increment can be called complete.
SFP: Small form factor pluggables.
shared storage: Storage that is configured so that multiple hosts can concurrently access the storage. The storage has a uniform appearance to all hosts. The host programs that access the storage must have a common model for the information on a storage device. The programs must be designed to handle the effects of concurrent access.
shortwave laser adapter: A connector that is used between host and DS6000 to support shortwave fibre-channel communication.
SIM: See service information message.
Simple Network Management Protocol (SNMP): In the Internet suite of protocols, a network management protocol that is used to monitor routers and attached networks. SNMP is an application layer protocol. Information on devices managed is defined and stored in the application's Management Information Base (MIB). (GC) See also management information base.
simplex volume: A volume that is not part of a FlashCopy, XRC, or PPRC volume pair.
small computer system interface (SCSI): A standard hardware interface that enables a variety of peripheral devices to communicate with one another. (GC)
smart relay host: A mail relay or mail gateway that has the capability to correct e-mail addressing problems.
SMIT: See System Management Interface Tool.
SMP: See symmetrical multiprocessor.
SNMP: See Simple Network Management Protocol.
SNMP agent: A server process that resides on a network node and is responsible for communicating with managers regarding that node. The node is represented as a managed object, which has various fields or variables that are defined in the appropriate MIB.
SNMP manager: A managing system that runs a managing application or suite of applications. These applications depend on Management Information Base (MIB) objects for information that resides on the managed system. Managers generate requests for this MIB information, and an SNMP agent on the managed system responds to these requests. A request can either be the retrieval or modification of MIB information.
software transparency: Criteria applied to a processing environment that states that changes do not require modifications to the host software in order to continue to provide an existing function.
source device: One of the devices in a dual-copy or remote-copy volume pair. All channel commands to the logical volume are directed to the source device. The data on the source device is duplicated on the target device. See also target device.
spare: A disk drive on the DS6000 that can replace a failed disk drive. A spare can be predesignated to allow automatic dynamic sparing. Any data preexisting on a disk drive that is invoked as a spare is destroyed by the dynamic sparing copy process.
spatial reuse: A feature of serial storage architecture that enables a device adapter loop to support many simultaneous read/write operations. See also serial storage architecture.
SSID: See subsystem identifier.
SSR: See service support representative.
stacked status: For zSeries, the condition when the control unit is in a holding status for the channel, and the last time the control unit attempted to present the status, the channel responded with the stack-status control.
stage operation: The operation of reading data from the physical disk drive into the cache.
staging: To move data from an offline or low-priority device back to an online or higher priority device, usually on demand of the system or on request of the user.
standard volume: A volume that emulates one of several zSeries volume types, including 3390-2, 3390-3, 3390-9, 3390-2 (3380-track mode), or 3390-3 (3380-track mode), by presenting the same number of cylinders and capacity to the host as provided by the native zSeries volume type of the same name.
STI: See self-timed interface.
storage area network: A network that connects a company's heterogeneous storage resources.
storage capacity: The amount of data that a storage medium can hold; usually expressed in kilobytes, megabytes, or gigabytes.
storage complex: A configuration of one or more storage units that is managed by a management console.
storage device: A physical unit that provides a mechanism to store data on a given medium such that it can be subsequently retrieved. See also disk drive module.
storage extent: The minimum contiguous range of storage on a physical storage device, array, or rank that can be allocated to a local volume
storage image: A partitioning of a storage unit that provides emulation of a storage server with one or more storage devices that provides storage capability to a host computer. You can configure more than one storage image on a storage unit. (DS8000 series only)
storage server: A physical unit that manages attached storage devices and provides an interface between them and a host computer by providing the function of one or more logical subsystems. The storage server can provide functions that the storage device does not provide. The storage server has one or more clusters.
storage unit: A physical unit that consists of a storage server that is integrated with one or more storage devices that provide storage capability to a host computer.
storage unit identifier: A unique identifier for a storage unit that consists of a manufacturer, a model number, a type number, a plant of manufacture, and a sequence number.
striping: A technique that distributes data in bit, byte, multibyte, record, or block increments across multiple disk drives.
subagent: An extension to an SNMP agent that permits a user to dynamically add, or in some cases replace, additional management variables in the local MIB, thereby providing a means of extending the range of information that network managers can access. See also agent.
subchannel: A logical function of a channel subsystem associated with the management of a single device.
subordinate storage unit: The physical unit that receives commands from the master storage unit and is specified when a Global Mirror session is started. The subordinate storage unit forms consistency groups and performs other Global Mirror processing. A subordinate storage unit can be controlled by only one master storage unit. Contrast with master storage unit.
subsystem identifier (SSID): A number that uniquely identifies a logical subsystem within a computer installation.
support catcher: See catcher.

support catcher telephone number: The telephone number that connects the support catcher server to the DS6000 to receive a trace or dump package. See also support catcher and remote technical assistance information network.
switched fabric: A fibre-channel topology in which ports are interconnected through a switch. Fabric switches can also be interconnected to support numerous ports on a single network. See also arbitrated loop and point-to-point connection.
symmetrical multiprocessor (SMP): An implementation of a multiprocessor computer consisting of several identical processors configured in a way that any subset of the set of processors is capable of continuing the operation of the computer. TheDS6000 contains four processors set up in SMP mode.
synchronous write: A write operation whose completion is indicated after the data has been stored on a storage device.
System/390: See S/390.
system adapter identification number (SAID): The unique identification number that is automatically assigned to each DS6000 host adapter for use by Copy Services.
System Management Interface Tool (SMIT): An interface tool of the AIX operating system for installing, maintaining, configuring, and diagnosing tasks.
System Modification Program: A program used to install software and software changes on MVS systems.

T

target: A SCSI device that acts as a subordinate to an initiator and consists of a set of one or more logical units, each with an assigned logical unit number (LUN). The logical units on the target are typically I/O devices. A SCSI target is analogous to a zSeries control unit. See also small computer system interface.
target device: One of the devices in a dual-copy or remote-copy volume pair that contains a duplicate of the data that is on the source device. Unlike the source device, the target device might only accept a limited subset of data. See also source device.
TB: See terabyte.
TCP/IP: See Transmission Control Protocol/Internet Protocol.
terabyte (TB): 1) Nominally, 1 000 000 000 000 bytes, which is accurate when speaking of bandwidth and disk storage capacity.; 2) For DS6000 cache memory, processor storage, real and virtual storage, a terabyte refers to 2⁴⁰ or 1 099 511 627 776 bytes.
terminal emulator: A function of the management console that allows it to emulate a terminal.
thousands of power-on hours (KPOH): A unit of time used to measure the mean time between failures (MTBF).
time sharing option (TSO): An operating system option that provides interactive time sharing from remote terminals.
System Storage: See IBM System Storage.
TPF: See transaction processing facility.
track: A unit of storage on a CKD device that can be formatted to contain a number of data records. See also home address, track-descriptor record, and data record.
track-descriptor record (R0): A special record on a track that follows the home address. The control program uses it to maintain certain information about the track. The record has a count field with a key length of zero, a data length of 8, and a record number of 0. This record is sometimes referred to as R0.
transaction processing facility (TPF): A high-availability, high-performance IBM operating system, designed to support real-time, transaction-driven applications. The specialized architecture of TPF is intended to optimize system efficiency, reliability, and responsiveness for data communication and database processing. TPF provides real-time inquiry and updates to a large, centralized database, where message length is relatively short in both directions, and response time is generally less than three seconds. Formerly known as the Airline Control Program/Transaction Processing Facility (ACP/TPF).
Transmission Control Protocol (TCP): A communications protocol used in the Internet and in any network that follows the Internet Engineering Task Force (IETF) standards for internetwork protocol. TCP provides a reliable host-to-host protocol between hosts in packet-switched communications networks and in interconnected systems of such networks. It uses the Internet Protocol (IP) as the underlying protocol.
Transmission Control Protocol/Internet Protocol (TCP/IP): 1) A combination of data-transmission protocols that provide end-to-end connections between applications over interconnected networks of different types.; 2) A suite of transport and application protocols that run over the Internet Protocol. (GC) See also Internet Protocol and Transmission Control Protocol.
transparency: See software transparency.
TSO: See time sharing option.
turbo processor: A faster multiprocessor that has six processors with common access to the main storage.

U

UFS: UNIX filing system.
Ultra-SCSI: An enhanced small computer system interface.
unconfigure: To delete the configuration.
unit address: For zSeries, the address associated with a device on a given control unit. On OEMI interfaces, the unit address specifies a control unit and device pair on the interface.
unprotected volume: An AS/400 term that indicates that the AS/400 host recognizes the volume as an unprotected device, even though the storage resides on a RAID-formatted array and is, therefore, fault tolerant by definition. The data in an unprotected volume can be mirrored. Also referred to as an unprotected device.
upper-layer protocol: The layer of the Internet Protocol (IP) that supports one or more logical protocols (for example, a SCSI-command protocol and an ESA/390 command protocol). Refer to ANSI X3.230-199x.
UTC: See Coordinated Universal Time.

V

virtual machine facility: A virtual data processing machine that appears to the user to be for the exclusive use of that user, but whose functions are accomplished by sharing the resources of a shared data processing system. An alternate name for the VM/370 IBM operating system.
vital product data (VPD): Information that uniquely defines the system, hardware, software, and microcode elements of a processing system.
VM: The root name of several IBM operating systems, such as VM/XA, VM/ESA, VM/CMS, and z/VM. See also virtual machine facility.
volume: For zSeries, the information recorded on a single unit of recording medium. Indirectly, it can refer to the unit of recording medium itself. On a nonremovable-medium storage device, the term can also indirectly refer to the storage device associated with the volume. When multiple volumes are stored on a single storage medium transparently to the program, the volumes can be referred to as logical volumes.
volume group: A collection of either physical or logical volumes.
volume label: A unique identifier that a user assigns to a logical volume.
VPD: See vital product data.
VSE/ESA: An IBM operating system, the letters of which represent virtual storage extended/enterprise systems architecture.

W

weight distribution area: The area that is required to distribute the weight of the storage unit.
worldwide node name (WWNN): A unique 64-bit identifier for a host that contains a fibre-channel port. See also worldwide port name.
worldwide port name (WWPN): A unique 64-bit identifier associated with a fibre-channel adapter port. It is assigned in an implementation- and protocol-independent manner. See also worldwide node name
write hit: A write operation in which the requested data is in the cache.
write penalty: The performance impact of a classical RAID-5 write operation.
WWNN: See worldwide node name.
WWPN: See worldwide port name.

X

xSeries: The product name of an IBM eServer product that emphasizes industry-standard server scalability and self-managing server technologies. It is the successor to the Netfinity family of servers.

Z

z/Architecture: An IBM architecture for mainframe computers and peripherals. The IBM eServer zSeries family of servers uses the z/Architecture architecture. It is the successor to the S/390 and 9672 family of servers. See also iSeries.
zoning: In fibre-channel environments, the grouping of multiple ports to form a virtual, private, storage network. Ports that are members of a zone can communicate with each other, but are isolated from ports in other zones.
z/OS: An operating system for the IBM eServer product line that supports 64-bit real storage.
z/OS Global Mirror: A function of a storage server that assists a control program to maintain a consistent copy of a logical volume on another storage unit. All modifications of the primary logical volume by any attached host are presented in order to a single host. The host then makes these modifications on the secondary logical volume. This function was formerly called extended remote copy or XRC.
zSeries: An IBM eServer family of servers that emphasizes near-zero downtime.; IBM enterprise servers based on z/Architecture.
zSeries storage: Storage arrays and logical volumes that are defined in the DS6000 as connected to zSeries servers.