Glossary of Terms

(All) level

The optional highest level of a dimension. The (All) level contains a single member that is the summary of all members of the immediately subordinate level.

acknowledged delivery

A positive confirmation from either the system or the recipient to SQL Server Notification Services that a notification has been delivered to the intended recipient.

active statement

An SQL statement that has been run but whose result set has not yet been canceled or fully processed.

ActiveX Data Objects (Multidimensional) (ADO MD)

A high-level, language-independent set of object-based data access interfaces optimized for multidimensional data applications. Visual Basic and other automation languages use ADO MD as the data access interface to multidimensional data storage. ADO MD is a part of ADO 2.0 and later versions.

add-in

A custom extension, written in any language that supports the Component Object Model (COM), that interacts with Analysis Manager and provides specific functionality. The most common language that is used for add-ins is Visual Basic. Add-ins are registered with the Analysis Add-in Manager. They are called by the Analysis Add-in Manager in response to user actions in the user interface.

adornment

A control or status area that is attached to the edge of a pane or window, such as a toolbar or ruler.

aggregate function

A function that performs a calculation on a column in a set of rows and returns a single value.

aggregation

1. A table or structure that contains precalculated data for a cube. 2. A collection of objects that makes a whole. An aggregation can be a concrete or conceptual set of whole-part relationships among objects.

aggregation usage

A cube attribute property that determines whether an attribute will be considered a candidate for an aggregation.

alias

An alternative name for a table or column in expressions that is often used to shorten the name for subsequent reference in code, prevent possible ambiguous references, or provide a more descriptive name in the query output. An alias can also be an alternative name for a server.

allocation unit

A set of pages that can be operated on as a whole. Pages that belong to an allocation unit are tracked by Index Allocation Map (IAM) pages. An allocation unit is made up of the IAM-page chain and all the pages marked as allocated in that IAM page chain. An allocation unit can contain only one IAM chain, and an IAM chain must belong to only one allocation unit.

ancestor

A member in a superior level in a dimension hierarchy that is related through lineage to the current member within the dimension hierarchy.

anonymous subscription

A type of pull subscription for which detailed information about the subscription and the Subscriber is not stored.

API server cursor

A server cursor that is built to support the cursor functions of an API, such as ODBC, OLE DB, ADO, and DB-Library. An application does not usually request a server cursor directly. Instead, it calls the cursor functions of the API. The SQL Server interface for that API implements a server cursor if that is the best way to support the requested cursor functionality.

application database

The database that stores user and system data for one application. The application database contains the events, subscriptions, and notifications for the application. It also contains system metadata, including the schemas for the events, subscriptions, and notifications for the application; and the match rules for the application.

application programming interface

A set of routines available in an application, such as ADO, for use by software programmers when designing an application interface.

archive file

The .cab file created by archiving an Analysis Services database.

article

Identifies a database object that is included in a publication. A publication can contain different types of articles, including tables, views, stored procedures, and other objects. When tables are published as articles, filters can be used to restrict the columns and rows of the data sent to Subscribers.

atomic

Either all of the transaction data modifications are performed or none of them are performed.

attribute hierarchy

A hierarchy of attribute members that contains the following levels: A leaf level that contains each distinct attribute member, with each member of the leaf level also known as a leaf member; intermediate levels if the attribute hierarchy is a parent-child hierarchy; and an optional (All) level that contains the aggregated value of the attribute hierarchy's leaf members, with the member of the (All) level also known as the (All) member. By default, an attribute hierarchy is defined for each dimension attribute. Attribute hierarchies are visible by default.

automatic failover

A failover that is initiated by the mirror server after the principal server is disconnected from both the mirror server and the witness. Assuming that the mirror database is already synchronized with the principal database, the mirror server takes over the role of principal server and brings its copy of the database online as the principal database.

autonomy

The independence of one site from other sites when performing modifications to data.

back up

To create a backup copy of a database, filegroup, file, or transaction log.

backup device

A tape or disk drive that contains backup media.

backup media

Disk files or tapes that are used to hold one or more backups.

balanced hierarchy

A dimension hierarchy in which all leaf nodes are the same distance from the root node.

batch

1. A set of SQL statements submitted together and runas a group. A script is often a series of batches submitted one after the other. 2. In Reporting Services, a collection of SOAP method calls within a single transaction.

binary large object

A piece of binary data that has an exceptionally large size, such as pictures or audio tracks that are stored as digital data, or any variable or table column that is large enough to hold such values. In Transact-SQL, a binary large object (BLOB) is stored in an image column. Sometimes the term BLOB is also applied to large character data values, such as those stored in text or ntext columns.

bitwise operation

An operation that manipulates a single bit, or tests whether a bit is on or off.

bookmark

A link in a report that jumps to another place within the report.

bootstrap

A program that installs or runs other programs.

built-in functions

A group of predefined functions provided as part of the Transact-SQL and Multidimensional Expressions (MDX) languages.

bulk import

To load a large set of data rows from a data file into a SQL Server table. The Database Engine applies logging and locking optimizations when it is possible.

business logic handler

A merge replication feature that allows you to run custom code during the synchronization process.

calculated column

A column in a table that displays the result of an expression rather than stored data. For example, CalculatedCostColumn = Price * Quantity.

calculated member

A member of a dimension whose value is calculated at run time by using an expression. Calculated member values can be derived from the values of other members. A calculated member is any member that is not an input member. For example, a calculated member, Profit, can be determined by subtracting the value of the member Costs, from the value of the member Sales.

calculation formula

A Multidimensional Expressions (MDX) expression used to supply a value for cells in a calculation subcube, subject to the application of a calculation condition.

calculation subcube

The set of multidimensional cube cells that is used to create a calculated cells definition. The set of cells is defined by a combination of Multidimensional Expressions (MDX) set expressions.

cascading delete

An operation that deletes a row containing a primary key value that is referenced by foreign key columns in existing rows in other tables. On a cascade delete, all of the rows whose foreign key values reference the deleted primary key value are also deleted.

cascading update

An operation that updates a primary key value that is referenced by foreign key columns in existing rows in other tables. On a cascade update, all the foreign key values are updated to match the new primary key value.

case key

In data mining, the element of a case by which the case is referenced within a case set.

cell

In a cube, the set of properties, including a value, specified by the intersection when one member is selected from each dimension.

change script

A text file that contains SQL statements for all changes made to a database, in the order in which they were made, during an editing session. Each change script is saved in a separate text file with an .sql extension. Change scripts can be applied back to the database later, using a tool such as osql.

character format

Data stored in a bulk copy data file using text characters.

chart data region

A report item on a report layout that displays data in a graphical format.

checkpoint

An event in which the Database Engine writes dirty buffer pages to disk. Each checkpoint writes to disk all the pages that were dirty at the last checkpoint and still have not been written to disk. Checkpoints occur periodically based on the number of log records that are generated by data modifications, or when checkpoints are requested by a user or a system shutdown.

child package

In Integration Services, a package that is run by another package. A child package is also known as the called package.

CLI

See Other Term: call-level interface

client application

An application that retrieves data from a server and performs local analysis and presentation of data from relational or multidimensional databases.

client subscription

A subscription to a merge publication that uses the priority value of the Publisher for conflict detection and resolution. Referred to as a local subscription in previous versions of SQL Server.

CLR function

A function that is created by referencing a SQL Server assembly. The implementation of the CLR function is defined in an assembly that is created in the .NET Framework common language runtime (CLR).

CLR trigger

A DML trigger or DDL trigger that is created by referencing a SQL Server assembly. The implementation of the CLR trigger is defined in an assembly that is created in the .NET Framework common language runtime (CLR).

clustered index

An index in which the logical order of the key values determines the physical order of the corresponding rows in a table. In a clustered index, the leaf nodes contain the data pages of the underlying table.

code access security

A mechanism provided by the common language runtime whereby managed code is granted permissions by security policy and these permissions are enforced, helping to limit the operations that the code will be allowed to perform.

collation

A set of rules that determines how data is compared, ordered, and presented. Character data is sorted using collation information, including locale, sort order, and case-sensitivity.

column

In an SQL table, the area in each row that stores the data value for some attribute of the object modeled by the table. For example, the Employee table in the AdventureWorks sample database models the employees of the Adventure Works Cycles company. The Title column in each row of the Employee table stores the job title of the employee represented by that row, the same way a Job Title field in a window or form would contain a job title.

column filter

Column filters restrict the columns to be included as part of a snapshot, transactional, or merge publication.

column-level constraint

A constraint definition that is specified within a column definition when a table is created or altered. The constraint applies only to the associated column.

command

In transactional replication, a command consists of one Transact-SQL data manipulation language (DML) statement or one data definition language (DDL) statement. Commands are stored in the distribution database and then delivered to Subscribers.

commit

An operation that saves all changes to databases, cubes, or dimensions made since the start of a transaction. A commit guarantees that all of the transaction's modifications are made a permanent part of the database, cube or dimension. A commit also frees resources, such as locks, used by the transaction.

complete database restore

A restore of a full database backup, the most recent differential database backup (if any), and the log backups (if any) taken since the full database backup.

Component Object Model

A Microsoft specification for developing component software. Several SQL Server and database APIs such as SQL-DMO, OLE DB, and ADO are based on COM. Some SQL Server components store objects as COM objects.

composite key

A key composed of two or more columns.

computed field

A value in a formatted notification that has been computed by using a Transact-SQL expression.

concurrency

A process that allows multiple users to access and change shared data at the same time. SQL Server uses locking to allow multiple users to access and change shared data at the same time without conflicting with each other.

configurations

In Integration Services, a name/value pair that updates the value of package objects when the package is loaded.

connection manager

In Integration Services, a logical representation of a run-time connection to a data source.

constraint

A property assigned to a table column that prevents certain types of invalid data values from being placed in the column. For example, a UNIQUE or PRIMARY KEY constraint prevents you from inserting a value that is a duplicate of an existing value, a CHECK constraint prevents you from inserting a value that does not match a specified condition, and NOT NULL prevents you from inserting a NULL value.

content formatter (Notification Services)

The part of the Distributor that turns raw notification data into readable messages.

control-of-flow language

Transact-SQL keywords that control the flow of execution of SQL statements and statement blocks in triggers, stored procedures, and batches.

CPU busy

A SQL Server statistic that reports the time, in milliseconds, that the central processing unit (CPU) spent on SQL Server work.

credential

Data used by a principal to establish the identity of the principal, such as a password or user name.

cube granularity

The space in a cube that is defined by the cross-join of a single level from each and every attribute hierarchy.

cursor

An entity that maps over a result set and establishes a position on a single row within the result set. After the cursor is positioned on a row, operations can be performed on that row, or on a block of rows starting at that position. The most common operation is to fetch (retrieve) the current row or block of rows.

cursor library

A part of the ODBC and DB-Library APIs that implements client cursors. A cursor library is not commonly used in current systems; server cursors are used instead.

custom rule

In a role, a specification that limits the dimension members or cube cells that users in the role are permitted to access.

data block

1. In text, ntext, and image data, a data block is the unit of data transferred at one time between an application and an instance of SQL Server. The term is also applied to the units of storage for these data types. 2. In tape backup files, data block is the unit of physical I/O.

data definition

Specifying the attributes, properties, and objects in a database.

data dictionary

A set of system tables, stored in a catalog, that includes definitions of database structures and related information, such as permissions.

data explosion

The exponential growth in size of a multidimensional structure, such as a cube, due to the storage of aggregated data.

data flow

The ordered workflow in an Integration Services package that extracts, transforms, and loads data.

data manipulation language

The subset of SQL statements that is used to retrieve and manipulate data. DML statements typically start with SELECT, INSERT, UPDATE, or DELETE.

data member

A child member associated with a parent member in a parent-child hierarchy. A data member contains the data value for its parent member, rather than the aggregated value for the parent's children.

data processing extension

A component in Reporting Services that is used to retrieve report data from an external data source. Data processing extensions can include special-purpose query designers that make it easier to work with data in Report Designer. Examples of some default data processing extensions include Microsoft SQL Server, Analysis Services, Oracle, and SAP NetWeaver Business Intelligence.

data region

A report item that displays repeated rows of data from an underlying dataset in a table, matrix, list, or chart.

data source

1. In ADO and OLE DB, the location of a source of data exposed by an OLE DB provider. 2. The source of data for an object such as a cube or dimension. It is also the specification of the information necessary to access source data. It sometimes refers to an object of ClassType clsDataSource. 3. In Reporting Services, a specified data source type, connection string, and credentials, which can be saved separately to a report server and shared among report projects or embedded in a report definition (.rdl) file.

data source view

A named, browsable, persisted selection of database objects that can be used as a data source by multiple projects.

data warehouse

A database specifically structured for query and analysis. A data warehouse typically contains data representing the business history of an organization.

database backup

A backup of every data file in a database. A full database backup contains all the data in every file.

database diagram

A graphical representation of the objects in a database. A database diagram can be either a whole or a partial picture of the structure of a database. It includes objects for tables, the columns that they contain, and the relationship between them.

database language

The language used for accessing, querying, updating, and managing data in relational database systems. SQL is a widely used database language. The Microsoft SQL Server implementation of SQL is called Transact-SQL.

database mirroring partners

Two server instances that act as role-switching partners for a mirrored database.

database owner

A member of the database administrator role of a database. There is only one database owner. The owner has full permissions in that database and determines the access and capabilities provided to other users.

database role

A collection of users and groups with the same access to an Analysis Services database. You can assign a database role to multiple cubes in the database, thereby granting the role's users access to these cubes.

database script

A collection of statements used to create database objects. Transact-SQL scripts are saved as files, usually ending with .sql.

data-definition query

An SQL query that contains data definition language (DDL) statements. These are statements that allow you to create or alter objects (such as tables, indexes, views, and so on) in the database and to migrate database objects from Microsoft Access.

datareader

A stream of data that is returned by an ADO.NET query. The DataReader object enables a client to receive a read-only, forward-only stream of data.

DBCS

See Other Term: double-byte character set

DCL

See Other Term: Data Control Language

DDL trigger

A trigger that fires in response to data definition language (DDL) statements.

decision support

Systems designed to support the complex analytic analysis required to discover business trends. The information retrieved from these systems allows managers to make business decisions based on timely and accurate analysis of business trends.

declarative referential integrity

FOREIGN KEY constraints defined as part of a table definition that enforce proper relationships between tables. The constraints ensure that proper actions are taken when DELETE, INSERT, and UPDATE statements remove, add, or modify primary or foreign key values. The DRI actions enforced by FOREIGN KEY constraints can be supplemented with additional referential integrity logic defined in triggers on a table.

DEFAULT constraint

A property defined for a table column that specifies a constant to be used as the default value for the column. If any subsequent INSERT or UPDATE statement specifies a value of NULL for the column, or does not specify a value for the column, the constant value defined in the DEFAULT constraint is placed in the column.

default instance

The instance of SQL Server that uses the same name as the computer name on which it is installed.

default member

The dimension member used in a query when no member is specified for the dimension. The default member of a dimension is the All member if an (All) level exists, or else an arbitrary member of the highest level. You can also set default members for individual roles in custom rules for dimension security.

deferred transaction

A transaction that is not committed when the roll forward phase of recovery finishes, and that cannot be rolled back during database startup because the data required by rollback is offline. This data can reside in either a page or a file.

delimiter

1. In Transact-SQL, characters that indicate the start and end of an object name, by using either double quotation marks ("") or brackets ([]). 2. In Integration Services and Reporting Services, characters that are used to separate records, fields, or strings.

delivery channel instance

A single occurrence of a delivery channel.

delivery extension

A component in Reporting Services that is used to distribute a report to specific devices or target locations. Examples of actual and potential delivery extensions include e-mail delivery, shared folder delivery, printer delivery, or application delivery (such as a document management system).

denormalize

To introduce redundancy into a table to incorporate data from a related table. The related table can then be eliminated. Denormalization can improve efficiency and performance by reducing complexity in a data warehouse schema.

dependencies

The objects that depend on other objects in the database. For example, a stored procedure might depend on a table, a view, or on another stored procedure.

destination

An Integration Services data flow component that writes the data from the data flow into a data source or creates an in-memory dataset.

device name

A user-defined description of a particular device, used in conjunction with the Subscriber ID to identify a specific Subscriber device. A device name must be provided to the notification function as an input parameter.

differential backup

A backup that contains only the changes that were made since the creation of its differential base or bases. A differential backup can be taken for a whole database or any subset of data files for which a base exists.

digest delivery (Notification Services)

A method of sending notifications that combines multiple notifications within a batch and sends the resulting message to a Subscriber.

dimension table

A table in a data warehouse whose entries describe data in a fact table. Dimension tables contain the data from which dimensions are created.

direct connect

The state of being connected to a back-end database, so that any changes you make to a database diagram automatically update your database when you save the diagram or selected items in it.

dirty pages

Buffer pages that contain modifications that have not been written to disk.

discretization

An automatic grouping of attribute hierarchy members.

distribute

To move transactions or snapshots of data from the Publisher to Subscribers, where they are applied to the destination tables in the subscription databases.

distributed query

A single query that accesses data from multiple data sources.

distribution database

A database on the Distributor that stores metadata and history data for all types of replication and transactions for transactional replication.

Distributor

A database instance that acts as a store for replication specific data associated with one or more Publishers. Each Publisher is associated with a single database (known as a distribution database) at the Distributor.

DMF

See Other Term: dynamic management function

DML trigger

A stored procedure that executes when data in a specified table is modified. Triggers are often created to enforce referential integrity or consistency among logically related data in different tables.

document map

A navigation pane in a report arranged in a hierarchy of links to report sections and groups.

domain integrity

An integrity mechanism that enforces the validity of entries for a given column. The mechanism, such as the CHECK constraint, can restrict the possible data values by data type, format, or range of values allowed.

download-only article

An article in a merge publication that can be updated only at the Publisher or at a Subscriber that uses a server subscription.

drill down/drill up

A technique for navigating through levels of data ranging from the most summarized (up) to the most detailed (down). For example, when viewing the details of sales data by year, a user can drill down to display sales data by quarter, and further to display data by month.

drillthrough report

A report with the 'enable drilldown' option selected. Drillthrough reports contain hyperlinks to related reports.

DSN-less connection

A type of data connection that is created based on information in a data source name (DSN), but is stored as part of a project or application. DSN-less connections are especially useful for Web applications because they let you move the application from one server to another without re-creating the DSN on the new server.

dynamic filter

See Other Term: parameterized row filter

dynamic management function

A set of built-in functions that return server state information about values, objects, and settings in SQL Server. These views that can be used to monitor the health of a server instance, diagnose problems, and tune performance.

dynamic recovery

The process that detects and/or attempts to correct software failure or loss of data integrity within a relational database management system (RDBMS).

dynamic SQL statements

In Embedded SQL for C, an SQL statement built and executed at run time.

entity integrity

A state in which every row of every table can be uniquely identified. Entity integrity is usually enforced by a PRIMARY KEY constraint, a UNIQUE constraint, or a UNIQUE index.

equijoin

A join in which the values in the columns being joined are compared for equality, and all columns are included in the results.

error state number

A number associated with SQL Server messages that helps Microsoft support engineers find the specific code location that issued the message. This can be helpful in diagnosing errors that might be generated from different locations in the SQL Server code.

ETL

Extraction, transformation, and loading. The complex process of copying and cleaning data from heterogeneous sources. Important part of development projects for data warehousing and business intelligence.

event batch (Notification Services)

A set of one or more events submitted to Notification Services at one time.

event chronicle rule

One or more Transact-SQL statements that manage the data in the event chronicle. This type of rule is fired each time an event batch for the rule's event class is submitted to the system.

event collection stored procedures

System-generated stored procedures that an application can call to submit events to the event table in the application database. These stored procedures can be called from SQL Server triggers or other SQL Server stored procedures.

event notification

A special kind of trigger that sends information about database events to a service broker.

event schema

An SQL-annotated XML Schema definition language (XSD) schema that maps event data to an event schema defined in the application definition file. Used by the EventLoader class.

event table

A table in the application database that stores event data. It is automatically generated based on information in the application definition file each time NSControl Create or NSControl Update is run. Each row in the table is an event instance.

exclusive lock

A lock that prevents any other transaction from acquiring a lock on a resource until the original lock on the resource is released at the end of the transaction. An exclusive lock is always applied during an update operation (INSERT, UPDATE, or DELETE).

execution tree

The path of data in the data flow of a SQL Server Integration Services package from sources through transformations to destinations. The scheduler of the data flow engine uses the trees to build the execution plan of the data flow.

expression

1. In SQL, a combination of symbols and operators that evaluate to a single data value. Simple expressions can be a constant, variable, column, or scalar function. Complex expressions are one or more simple expressions connected by operators. 2. In Integration Services, a combination of literals, constants, functions, and operators that evaluate to a single data value.

Extensible Stylesheet Language (XSL)

An XML vocabulary that is used to transform XML data to another form, such as HTML, by means of a style sheet that defines presentation rules.

extent

The unit of space allocated to a SQL Server object, such as a table or index, whenever the object needs more space.

fact

A row in a fact table in a data warehouse. A fact contains values that define a data event such as a sales transaction.

fact table

A central table in a data warehouse schema that contains numerical measures and keys relating facts to dimension tables. Fact tables contain data that describes specific events within a business, such as bank transactions or product sales.

federated database servers

A set of linked servers that shares the processing load of data by hosting partitions of a distributed partitioned view.

field

1. An area in a window or record that stores a single data value. Some databases, such as Access databases, use field as a synonym for column. 2. In Report Builder, a field is an attribute of the entity that is a named unit of data that can be added to a report layout template.

field terminator

In bulk copy, one or more characters marking the end of a field or row, separating one field or row in the data file from the next.

file backup

A backup of all the data in one or more files or filegroups.

file restore

An operation that restores one or more files of a database.

file system watcher event provider (Notification Services)

An event provider supplied by Notification Services that monitors a directory for the arrival of XML event files, extracts XML event data from these files, and submits these events to the event table in the application database. The file system watcher uses the event loader class to submit events.

fill factor

An attribute of an index that defines how full the SQL Server Database Engine should make each page of the index. The fill factor accommodates the future growth of data in a table and reduces the potential for page splits. The FILLFACTOR option can be a value from 1 up to 100. This value specifies the percentage of space on each leaf-level page to be filled with data.

filtering

The ability to restrict data based upon criteria set in the WHERE clause of an SQL statement. For replication, filtering occurs on table articles defined in a publication. The result is partitions of data that can be published to Subscribers.

FK

See Other Term: foreign key

flattened rowset

A multidimensional data set presented as a two-dimensional rowset in which unique combinations of elements of multiple dimensions are combined on an axis. For more information, see the OLE DB documentation.

foreign key

The column or combination of columns whose values match the primary key (PK) or unique key in the same or another table. Also referred to as the referencing key.

forward-only cursor

A cursor that cannot be scrolled; rows can be read only in sequence from the first row to the last row.

full outer join

A type of outer join in which all rows in all joined tables are included, whether they are matched or not.

full-text catalog

Stores all of the full-text indexes for tables within a database.

full-text query

As a SELECT statement, a query that searches for words, phrases, or multiple forms of a word or phrase in the character-based columns (of char, varchar, text, ntext, nchar, or nvarchar data types). The SELECT statement returns those rows meeting the search criteria.

function

A piece of code that operates as a single logical unit. A function is called by name, accepts optional input parameters, and returns a status and optional output parameters. Many programming languages support functions, including C, Visual Basic, and Transact-SQL. Transact-SQL supplies built-in functions, which cannot be modified, and supports user-defined functions, which can be created and modified by users.

fuzzy matching

In Integration Services, a lookup methodology that uses an approximate matching algorithm to locate similar data values in a reference table.

global default

A default that is defined for a specific database and is shared by columns of different tables.

global subscription

See Other Term: server subscription

granularity attribute

The attribute of a cube dimension that links a dimension to the facts in a measure group in the measure's dimension. If the granularity attribute and the key attribute are different attributes, then non-key attributes must be linked, directly or indirectly, to the granularity attribute. Within a cube, the granularity attribute defines a dimension's granularity.

heap

A table without a clustered index.

hierarchy

A logical tree structure that organizes the members of a dimension such that each member has one parent member and zero or more child members.

homogeneous data

Data that comes from multiple data sources that are all managed by the same software. A SQL Server distributed query is homogeneous if all the data comes from SQL Server instances.

horizontal partitioning

To segment a single table into multiple tables based on selected rows. Each of the multiple tables has the same columns but fewer rows.

hot standby server

A standby server that can support rapid failover without a loss of data from committed transactions. Database mirroring can be used to maintain a hot standby server, depending on the configuration and state of the mirroring session. When the session is synchronized, the mirror server provides a hot standby server.

Hypertext Markup Language (HTML)

A system of marking up, or tagging, a document so that it can be published on the World Wide Web. Documents prepared in HTML include reference graphics and formatting tags. You use a Web browser (such as Microsoft Internet Explorer) to view these documents.

identifier

The name of an object in a database. An identifier can be from 1 through 128 characters.

identity column

A column in a table that has been assigned the identity property. The identity property generates unique numbers.

idle time

A SQL Server Agent condition that defines how much CPU usage by the SQL Server Database Engine constitutes an idle state. SQL Server Agent jobs can then be created to run whenever the Database Engine CPU usage falls below the level that is defined in the idle time definition. This minimizes the effect the SQL Server Agent jobs might have on other tasks that are accessing the database.

immediate updating subscriptions

A subscription to a transactional publication for which the user is able to make data modifications at the Subscriber. The data modifications are then immediately propagated to the Publisher using the two-phase commit protocol (2PC).

implicit transaction

A connection option in which each SQL statement executed by the connection is considered a separate transaction.

incremental update

The set of operations that either adds new members to an existing cube or dimension, or adds new data to a partition. An incremental update is one of three processing options for a cube or partition, and one of two processing options for a dimension.

index

In a relational database, a database object that provides fast access to data in the rows of a table, based on key values. Indexes can also enforce uniqueness on the rows in a table. SQL Server supports clustered and nonclustered indexes. The primary key of a table is automatically indexed. In full-text search, a full-text index stores information about significant words and their location within a given column.

index ORing

An execution strategy that consists of looking up rows of a single table using several indexes, followed by producing the result (by combining the partial results). For example, WHERE R.a = 6 OR R.b = 7 with indexes on columns R.a and R.b.

indexed view

A view that has a unique clustered index applied on it. Indexed views can improve the performance of some types of queries.

initial snapshot

Files that include schema and data, constraints, extended properties, indexes, triggers, and system tables that are necessary for replication. By default, the initial snapshot is transferred to Subscribers when implementing replication.

initial tape

In a media set using tape backup devices, the first tape in a media family.

input member

A member whose value is loaded directly from the data source instead of being calculated from other data.

input source

Any table, view, or schema diagram used as an information source for a query.

Insert Into query

A query that copies specific columns and rows from one table to another or to the same table.

instance

A copy of SQL Server that is running on a computer.

integrity constraint

A property defined on a table that prevents data modifications that would create invalid data.

interactive structured query language

An interactive command prompt utility provided with SQL Server that lets users run Transact-SQL statements or batches from a server or workstation and view the results that are returned.

interface implication

If an interface implies another interface, then any class that implements the first interface must also implement the second interface. Interface implication is used in an information model to get some of the effects of multiple inheritance.

Internet Protocol security

Rules that computers follow to provide private and secure communication over Internet Protocol (IP) networks, through the use of cryptographic security services.

interprocess communication

A mechanism through which operating system processes and threads exchange data and messages. IPCs include local mechanisms such as Windows shared memory, or network mechanisms such as Windows Sockets.

IPsec

See Other Term: Internet Protocol security

ISQL

See Other Term: interactive structured query language

join

As a verb, to combine the contents of two or more tables and produce a result set that incorporates rows and columns from each table. Tables are typically joined using data that they have in common. As a noun, the process or result of joining tables, as in the term "inner join" to indicate a particular method of joining tables.

join condition

A comparison clause that specifies how tables are related by their join columns.

join operator

A comparison operator in a join condition that determines how the two sides of the condition are evaluated and which rows are returned.

junction table

A table that establishes a relationship between other tables. The junction table contains foreign keys referencing the tables that form the relationship. For example, an OrderParts junction table can show what parts shipped with each order by having foreign keys to an Orders table and a Parts table.

key

A column or group of columns that uniquely identifies a row (primary key), defines the relationship between two tables (foreign key), or is used to build an index.

key column

1. In an Analysis Services dimension, an attribute property that uniquely identifies the attribute members. 2. In an Analysis Services mining model, a data mining column that uniquely identifies each case in a case table.

key range lock

A lock that is used to lock ranges between records in a table to prevent phantom additions to, or deletions from, a set of records. Ensures serializable transactions.

keyword

A reserved word in SQL Server that performs a specific function, such as to define, manipulate, or access database objects.

latency

The amount of time that elapses when a data change is completed at one server and when that change appears at another (for example, the time between when a change is made at a Publisher and when it appears at the Subscriber).

leaf level

The bottom level of a clustered or nonclustered index. In a clustered index, the leaf level contains the actual data pages of the table. In a nonclustered index, the leaf level either points to data pages or points to the clustered index (if one exists), rather than containing the data itself. In Analysis Services, the bottom level of a hierarchy.

left outer join

A type of outer join in which all rows from the left-most table in the JOIN clause are included. When rows in the left table are not matched by rows in the right table, all result set columns that come from the right table are assigned a value of NULL.

level naming template

A template used to construct level names in a parent-child hierarchy.

linguistic casing

Functionality provided by the Win32 API for Unicode simple case mapping of Turkic and other locales.

linked measure

A measure based on a measure defined in another Analysis Services database.

linked server

A definition of an OLE DB data source used by SQL Server distributed queries. The linked server definition specifies the OLE DB provider required to access the data, and includes enough addressing information for the OLE DB provider to connect to the data. Any rowsets exposed by the OLE DB data source can then be referenced as tables, called linked tables, in SQL Server distributed queries.

linking table

A table that has associations with two other tables, and is used indirectly as an association between those two tables.

local cube

A cube created and stored with the extension .cub on a local computer using PivotTable Service.

local group

A group in Windows NT 4.0 or Windows 2000 containing user accounts and global groups from the domain group in which they are created and any trusted domain. Local groups cannot contain other local groups.

local partitioned view

A view that joins horizontally partitioned data from a set of member tables that exist on one server. The data appears as if it is in one table.

local subscription

See Other Term: client subscription

locale

The Windows operating system attribute that defines certain behaviors related to language. The locale defines the code page, or bit patterns, used to store character data, and the order in which characters are sorted. It also defines language-specific items such as the format used for dates and time and the character used to separate decimals in numbers. Each locale is identified by a unique number, referred to as a locale identifier or LCID. SQL Server collations are similar to locales in that the collations define language-specific types of behaviors for instances of SQL Server.

lock

A restriction on access to a resource in a multiuser environment. SQL Server locks users out of a specific row, column, or file automatically to maintain security or prevent concurrent data modification problems.

log backup

A backup of transaction logs that includes all log records that were not backed up in previous log backups. Log backups are required under the full and bulk-logged recovery models. Log backups are unavailable under the simple recovery model.

log file

A file or set of files that contain records of the modifications made in a database.

log sequence number

A number that uniquely identifies every record in the Microsoft SQL Server transaction log, in the order in which the underlying transaction occurred.

log shipping configuration

A configuration of servers that includes one primary server, one or more secondary servers (each with a secondary database), and a monitor server.

logical backup device

An optional user-defined name that points to a specific physical backup device, such as a disk file or tape drive.

logical operators

The operators AND, OR, and NOT. Used to connect search conditions in WHERE clauses.

login security mode

A security mode that determines the manner in which an instance of SQL Server validates a login request. There are three types of login security: Windows authentication, SQL Server authentication, and Mixed Mode authentication. Mixed Mode combines the Windows and SQL Server authentication modes.

LSN

See Other Term: log sequence number

Make Table query

A query, or SQL statement, that creates a new table, and then creates rows in the table by copying rows from an existing table.

many-to-many relationship

A relationship between two tables in which rows in each table have multiple matching rows in the related table. Many-to-many relationships are maintained by using a third table, referred to as a junction table, and adding the primary key columns from each of the other two tables to this table.

master database

The system database that records all the system-level information for an instance of SQL Server. This includes instance-wide metadata such as login accounts, endpoints, linked servers, and system configuration settings. Also, master is the database that records the existence of all other databases and the location of those database files and records the initialization information for SQL Server.

matrix data region

A report item on a report layout that displays data in a variable columnar format.

measure

In a cube, a set of values that are usually numeric and are based on a column in the fact table of the cube. Measures are the central values that are aggregated and analyzed.

measure group

A collection of related measures in an Analysis Services cube. The measures are generally from the same fact table.

measures dimension

The dimension that contains all the measures in a cube.

media family

Data written by a backup operation to a backup device that is being used by a media set. In a media set that has one device, only one media family exists. In a striped media set, multiple media families exist. If the striped media set is unmirrored, each device corresponds to a family. If the striped media set is mirrored, the mirrored media set contains from two to four identical copies of each media family, which are called mirrors.

media name

The descriptive name for the whole backup media set.

member

An item in a dimension representing one or more occurrences of data. A member can be either unique or nonunique. For example, 1997 and 1998 represent unique members in the year level of a time dimension, whereas January represents nonunique members in the month level because there can be more than one January in the time dimension if it contains data for more than one year.

memo

A type of column that contains long strings, typically more than 255 characters. A memo is the equivalent in Microsoft Access to a SQL Server text data type.

merge replication

A type of replication that typically starts with a snapshot of the publication database objects and data. Subsequent data changes and schema modifications made at the Publisher and Subscribers are tracked with triggers. The Subscriber synchronizes with the Publisher when connected to the network and exchanges all rows that have changed between the Publisher and Subscriber since the last time synchronization occurred.

Message Queuing

An asynchronous messaging framework distributed with Windows Server.

method

A function that performs an action by using a COM object, as in SQL-DMO, OLE DB, and ActiveX Data Objects (ADO).

mining model training

The process a data mining model uses to estimate model parameters by evaluating a set of known and predictable data. Also, the act of causing a mining model to evaluate training data.

mirror server

In a database mirroring configuration, the server instance on which the mirror database resides.

mirroring

The process that helps protect against the loss of data caused by disk failure by maintaining a fully redundant copy of data on a separate disk. Mirroring can be implemented at several levels: in SQL Server, in the operating system, and in the disk controller hardware.

model database

A database installed with SQL Server that provides the template for new user databases. SQL Server creates a new database by copying in the contents of the model database, and then expanding it to the size requested.

module

A group of objects in a project. You can move objects between modules in a project, thus organizing those objects for a dispersed development environment.

monitor server

In a log shipping configuration, a server instance on which every log shipping job in the configuration records its history and status. A single-monitor server can monitor multiple log shipping configurations.

MSX

See Other Term: master server

multicast delivery

A method for delivering notifications that formats a notification once, and then sends the resulting message to multiple Subscribers.

multidimensional OLAP

A storage mode that uses a proprietary multidimensional structure to store a partition's facts and aggregations or a dimension. The data of a partition is completely contained within the multidimensional structure.

multiple inheritance

A modeling term that describes how an interface receives the characteristics of more than one parent interface.

multithreaded server application

An application that creates multiple threads within a single process to service multiple user requests at the same time.

name column

The property of an attribute that specifies the names of the attribute members that appear to users.

named pipe

An interprocess communication (IPC) mechanism that SQL Server uses to provide communication between clients and servers. Named pipes permit access to shared network resources.

naming relationship

A naming convention that identifies the destination objects of that relationship by name.

natural hierarchy

A user-defined hierarchy in which a one-to-many relationship is defined between levels.

nested table

A data mining model configuration in which a column of a table contains a table.

nickname

When used with merge replication system tables, a name for another Subscriber that is known to already have a specified generation of updated data. Used to avoid sending an update to a Subscriber that has already received those changes.

noise word

Words that do not participate in a full-text query search, such as "a", "and", and "the".

non-empty behavior

A property of a calculated member, which is a base member; if the base member is null, Analysis Services will not resolve the calculated member.

nonleaf member

A member with one or more descendants.

non-uniform memory access

A multiprocessor architecture that divides a system into nodes. Each node might include one or more processors, some memory, and a bus. A node's memory is "local" to its processors, and the processors can access the memory relatively fast. Processors in one node can also access memory in different ("distant") nodes, but access time is slower. "Non-uniform memory access" refers to this difference between the speed at which processors can access local memory and the speed at which they can access distant memory. Most NUMA architectures are also ccNUMA (cache-coherent NUMA) architectures, because they provide cache coherency across the nodes.

NS$instance_name (Notification Services)

The name of the Windows service that runs an instance of Notification Services. Note that instance_name is a placeholder for the actual instance name, such as NS$ThisInstance.

nullability

The attribute of a column, parameter, or variable that specifies whether it allows null data values.

numeric expression

Any expression that evaluates to a number. The expression can be any combination of variables, constants, functions, and operators.

object identifier

1. A unique name given to an object. 2. In Metadata Services, a unique identifier constructed from a globally unique identifier (GUID) and an internal identifier. All objects must have an object identifier.

ODBC

See Other Term: Open Database Connectivity

ODBC driver

A dynamic-link library (DLL) that an ODBC-enabled application, such as Excel, can use to access an ODBC data source. Each ODBC driver is specific to a database management system (DBMS), such as SQL Server or Access.

OLAP

See Other Term: online analytical processing (OLAP)

OLE Automation objects

A Component Object Model (COM) object that provides Automation-compatible interfaces.

OLE DB

A COM-based API for accessing data. OLE DB supports accessing data stored in any format (databases, spreadsheets, text files, and so on) for which an OLE DB provider is available.

OLE DB for OLAP

Formerly, the separate specification that addressed OLAP extensions to OLE DB. Beginning with OLE DB 2.0, OLAP extensions are incorporated into the OLE DB specification.

OLTP

See Other Term: online transaction processing

one-to-one relationship

In relational database, a relationship between two tables in which a single row in the first table can be related only to one row in the second table, and a row in the second table can be related only to one row in the first table.

online mode

A direct connection to an Analysis Services database, through which changes can be made directly to that database rather than through a project.

online transaction processing

A data processing system designed to record all of the business transactions of an organization as they occur. An OLTP system is characterized by many concurrent users actively adding and modifying data.

Open Database Connectivity

A data access API that supports access to any data source for which an ODBC driver is available. ODBC is aligned with the American National Standards Institute (ANSI) and International Organization for Standardization (ISO) standards for a database call-level interface (CLI).

optimize synchronization

An option in merge replication that allows you to minimize network traffic when determining whether recent changes have caused a row to move into or out of a partition that is published to a Subscriber.

origin object

An object in a repository that is the origin in a directional relationship.

overfitting

The characteristic of some data mining algorithms that assigns importance to random variations in data by viewing them as important patterns.

page

In a virtual storage system, a fixed-length block of contiguous virtual addresses copied as a unit from memory to disk and back during paging operations. SQL Server allocates database space in pages. In SQL Server, a page is 8 KB in size.

page header

A set of static text, images, lines, rectangles, borders, background color, and background images that repeats on the top of each page of a report.

page split

The process of moving half the rows or entries in a full data or index page to a new page to make room for a new row or index entry.

parameterized report

A published Reporting Services report that accepts input values through parameters.

parent

A member in the next higher level in a hierarchy that is directly related to the current member. The parent value is usually a consolidation of the values of all of its children. For example, in a Time dimension containing the levels Quarter, Month, and Day, Qtr1 is the parent of January.

parent-child hierarchy

A special type of attribute hierarchy in which an attribute in the dimension is set to parent. A parent-child hierarchy is an unbalanced hierarchy of child and parent members. A parent-child hierarchy contains the following levels: child levels that contain the children of parent members (The children of a parent include the attribute members that aggregate to the parent member, including data members.); Intermediate levels that contain parent members; and an optional (All) level that contains the aggregated value of the parent-child hierarchies leaf members, with the member of the (All) level also known as the (All) member. Only one parent-child hierarchy can exist per dimension and must be related to the key attribute.

partial database restore

A restore of only a part of a database that consists of its primary filegroup and, perhaps, one or more secondary filegroups. Initially, the other filegroups remain offline, but they can be restored later.

partition function

A function that defines how the rows of a partitioned table or index are spread across a set of partitions based on the values of specific columns, which are referred to as partitioning columns

partitioned snapshot

In merge replication, a snapshot that includes only the data from a single partition. A partitioned snapshot is used to initialize subscriptions to publications that have parameterized filters. Referred to as a dynamic snapshot in previous versions of SQL Server.

partitioning

The process of replacing a table with multiple smaller tables. Each smaller table has the same format as the original table, but with a subset of the data. Each partitioned table has rows allocated to it based on some characteristic of the data, such as specific key ranges. The rules that define into which table the rows go must be unambiguous. For example, a table is partitioned into two tables. All rows with primary key values lower than a specified value are allocated to one table, and all keys equal to or greater than the value are allocated to the other. Partitioning can improve application processing speeds and reduce the potential for conflicts in multisite update replication. You can improve the usability of partitioned tables by creating a view. The view, created by a union of select operations on all the partitioned tables, presents the data as if it all resided in a single table.

partner

In database mirroring, refers to the principal server or the mirror server.

pass-through query

A query passed uninterpreted to an external server for evaluation. The result set returned by a pass-through query can be used in the FROM clause of a query like an ordinary base table.

password policy

A collection of policy settings that define the password requirements for a Group Policy object (GPO).

peer-to-peer replication

A type of transactional replication. In contrast to read-only transactional replication and transactional replication with updating subscriptions, the relationships between nodes in a peer-to-peer replication topology are peer relationships rather than hierarchical ones, with each node containing identical schema and data.

persistence

The saving of an object definition so it will be available after the current session ends.

phantom

By one task, the insertion of a new row or the deletion of an existing row in a range of rows previously read by another task that has not yet committed its transaction. The task with the uncommitted transaction cannot repeat its original read because of the change to the number of rows in the range. If a connection sets its transaction isolation level to serializable, SQL Server uses key-range locking to prevent phantoms.

physical name

The path where a file or mirrored file is located. The default is the path of the Master.dat file followed by the first eight characters of the file's logical name. For example, if Accounting is the logical name, and the Master.dat file is located in Sql\Data, the default physical name is Sql\Data\Accounti.dat. For a mirrored file, the default is the path of the Master.mir file followed by the first eight characters of the mirror file's logical name. For example, if Maccount is the name of the mirrored file, and the Master.mir file is located in Sql\Data, the default physical name is Sql\Data\Maccount.mir.

piecemeal restore

A composite restore process in which a database is restored in stages, where each stage corresponds to a restore sequence. The initial, partial restore sequence restores the files in the primary filegroup and perhaps other files to any point in time that is supported by the recovery model, and then brings the database online. Subsequent restore sequences bring remaining files to the point that is consistent with the database and then bring them online.

PK

See Other Term: primary key

plan guide

A SQL Server module that attaches query hints to queries in deployed applications, without directly modifying the query.

point-in-time recovery

The process of recovering only the transactions within a log backup that were committed before a specific point in time, instead of recovering the whole backup.

position

The current location of processing in a cursor. For example, after an application fetches the first 10 rows from a cursor, it is positioned on the tenth row of the cursor. Database APIs also have functions, such as the ODBC SQLSetPos function, that allow an application to move directly to a specific position in a cursor without performing a fetch.

precedence constraint

In Integration Services, precedence constraints connect executables, containers and tasks in packages, and specify conditions that determine whether executables run.

precomputed partitions

In merge replication, a performance optimization that is used by default for filtered publications.

prefix characters

A set of 1 to 4 bytes that prefix each data field in a native-format bulk-copy data file. The prefix characters record the length of the data value in the field, or contain -1 when the value is NULL.

prefix search

Full-text query searching for those columns where the specified character-based text, word, or phrase, is the prefix. When using a phrase, each word within the phrase is considered to be a prefix. For example, a prefix search specifying the phrase "sport fish*" matches "sport fishing", "sportsman fishing supplies", and so on.

primary dimension table

In a snowflake schema in a data warehouse, a dimension table that is directly related to and usually joined to the fact table. Additional tables that complete the dimension definition are joined to the primary dimension table instead of to the fact table.

primary server

In a log shipping configuration, the server instance on which the primary database resides.

principal database

In database mirroring, a read/write database whose transaction log is continuously sent to the mirror server. This restores the log to the mirror database.

procedure cache

The part of the SQL Server memory pool that is used to store execution plans for Transact-SQL batches, stored procedures, and triggers. When any SQL statement is executed, the relational engine first looks through the procedure cache to verify that an existing execution plan for the same SQL statement exists. Any existing plan is reused, saving the overhead of recompiling the SQL statement. If no existing execution plan exists, SQL Server generates a new execution plan for the query.

producer

Collects events in a specific event category and sends the data to a SQL Server Profiler queue.

property

A named attribute of a control, field, or database object that you set to define one of the object's characteristics, such as size, color, or screen location; or an aspect of its behavior, such as whether it is hidden.

provider

1. An OLE DB provider. 2. An in-process dynamic link library (DLL) that provides access to a database.

provision

To install, to supply, to grant, or to provide. For example: "You must provision the server with a certificate." or "The SQL Server login must be provisioned with CREATE DATABASE permission."

publication

A publication is a collection of one or more articles from one database. The grouping of multiple articles into a publication makes it easier to specify a logically related set of database objects and data that are replicated as a unit.

publication retention period

In merge replication, the amount of time a subscription can remain unsynchronized. If the retention period is exceeded, the subscription might require reinitialization. Also referred to as the subscription expiration period.

Publisher

A database instance that makes data available to other locations through replication. The Publisher can have one or more publications, each defining a logically related set of objects and data to replicate.

publishing table

The table at the Publisher in which data has been marked for replication and is part of a publication.

push subscription

A subscription created and administered at the Publisher. The Distribution Agent or Merge Agent for the subscription runs at the Distributor.

quantum

A period of time defined in the application definition file that sets the timing for generator firing.

query optimizer

The SQL Server Database Engine component responsible for generating efficient execution plans for SQL statements.

queue

A SQL Server Profiler queue that provides a temporary holding location for server events to be captured.

quorum

In a database mirroring session for which a witness server instance is set, a relationship established only when at least two of the server instances in the session are connected to each other. Quorum is required to make the database available.

range partitioning

A way of partitioning a table or index by specifying partitions to hold rows that have ranges of values from one partitioning column.

rank

For full-text and SQL Server Books Online searches, a value indicating how closely rows or topics match the specified search criteria. For Metadata Services and Analysis Services, a value indicating the relative positions of elements such as dimension members, hierarchy levels, or tuples in a set.

RDL

See Other Term: Report Definition Language

recordset

The ActiveX Database Objects (ADO) object that is used to contain a result set. It also exhibits cursor behavior depending on the recordset properties set by an application. ADO recordsets are mapped to OLE DB rowsets.

recovery branch

A range of LSNs that share the same recovery branch GUID. A new recovery branch originates when a database is created or when RESTORE WITH RECOVERY generates a recovery fork. A multiple-branch recovery path that includes ranges of LSNs that cover two or more recovery fork points is possible.

recovery interval

The maximum amount of time that the Database Engine should require to recover a database. The Database Engine makes sure that the active part of the database log is small enough to recover the database in the amount of time specified for the recovery interval.

recovery path

The sequence of data and log backups that have brought a database to a particular point in time, known as a recovery point. A recovery path is a specific set of transformations that have evolved the database over time, yet have maintained the consistency of the database. A recovery path describes a range of log sequence numbers. The range of log sequence numbers in a recovery path can traverse one or more recovery branches from start to end.

recursive hierarchy

A hierarchy of data in which all parent-child relationships are represented in the data.

redo queue

In database mirroring, copies of log records from the principal database that have been written to disk on the mirror server and are waiting to be applied to the mirror database.

referenced relationship

An indirect relationship between a dimension and a measure group based on another dimension.

referential integrity

A state in which all foreign key values in a database are valid. For a foreign key to be valid, it must contain either the value NULL, or an existing key value from the primary or unique key columns referenced by the foreign key.

refresh data

The series of operations that clears data from a cube, loads the cube with new data from the data warehouse, and calculates aggregations. Refresh data is used when a cube's underlying data in the data warehouse changes but the cube's structure and aggregation definitions remain the same. One of three processing options for a cube.

regular relationship

A direct relationship between a dimension and a measure group based on the key attribute.

relational database management system (RDBMS)

A system that organizes data into related rows and columns. SQL Server is a relational database management system (RDBMS).

relationship

1. A link between tables that references the primary key in one table to a foreign key in another table. The relationship line is represented in a database diagram by a solid line if referential integrity between the tables is enforced, or a dashed line if referential integrity is not enforced for INSERT and UPDATE transactions. The endpoints of a relationship line show a primary key symbol to denote a primary key-to-foreign key relationship, or they show an infinity symbol to denote the foreign key side of a one-to-many relationship. 2. In Metadata Services, a relationship is an association between a pair of objects, where one object is an origin and the other object is a destination. The association repeats for each subsequent pair of objects, so that the destination of one relationship becomes the origin in the next relationship. In this way, all objects in an information model are associated through a chain of relationships that extend from one object to the next throughout the information model.

relationship type

A definition of a relationship between two interfaces, as defined in an information model. A relationship type is similar to a class in that it describes characteristics to which specific instances must conform.

remote Distributor

A server configured as a Distributor that is separate from the server configured as the Publisher.

remote partition

A partition whose data is stored on a server running an instance of Analysis Services, other than the one used to store the metadata of the partition.

remote table

A table stored in an OLE DB data source that is separate from the current instance of SQL Server. The table is accessed by either establishing a linked server definition or using an ad-hoc connector name.

rendering extension

A component in Reporting Services that is used to process the output format of a report. Rendering extensions included in Reporting Services are HTML, TIFF, XML, Excel, PDF, CSV, and Web archive.

replication

A set of technologies for copying and distributing data and database objects from one database to another and then synchronizing between databases to maintain consistency.

Replication Monitor

Provides a systemic view of replication activity, focusing on the movement of data between the Publisher and the Subscribers. Replication Monitor is a tool for watching real-time activity, troubleshooting problems, and analyzing past replication activity.

replication topology

Defines the relationship between servers and copies of data and clarifies the logic that determines how data flows between servers.

Report Definition Language

A set of instructions that describe layout and query information for a report. RDL is composed of XML elements that conform to an XML grammar created for Reporting Services.

report item

Any object, such as a text box, graphical element, or data region, that exists on a report layout.

report layout template

A pre-designed table, matrix, or chart report template in Report Builder.

report processing extension

A component in Reporting Services that is used to extend the report processing logic. The Report Processor component provided with Reporting Services includes the ability to process a fixed set of report items (list, table, matrix, chart, textbox, line, rectangle, and image). Reporting Services also supports the ability to extend this list of report items through custom report items defined in report processing extensions. You can build or purchase a third-party report processing extension to support data-bound controls that you embed in your reports.

report rendering

The action of combining the report layout with the data from the data source for the purpose of viewing the report.

report server administrator

A user who is assigned to the Content Manager role, the System Administrator role, or both for a report server.

report server execution account

The account under which the Report Server Web service and Report Server Windows service run.

report snapshot

A report that contains data captured at a specific point in time. A report snapshot is actually a report definition that contains a dataset instead of query instructions.

repository

A database containing information models that, in conjunction with the executable software, manage the database.

republisher

A Subscriber that publishes data that it has received from a Publisher.

restore

A multiphase process that copies all the data and log pages from a specified backup to a specified database (the data-copy phase) and rolls forward all the transactions that are logged in the backup (the redo phase). At this point, by default, a restore rolls back any incomplete transactions (the undo phase). This completes the recovery of the database and makes it available to users.

result set

The set of rows returned from a SELECT statement. The format of the rows in the result set is defined by the column-list of the SELECT statement.

reusable bookmark

A bookmark that can be consumed from a rowset for a given table and used on a different rowset of the same table to position on a corresponding row.

RID

See Other Term: row identifier

ROLAP

See Other Term: relational OLAP

role assignment

In Reporting Services, a role assignment is a security policy that determines whether a user or group can access a specific item and perform an operation. A role assignment consists of a user or group account name and one or more role definitions.

role switching

In a database mirroring session, the taking over of the principal role by the mirror.

roll forward

To apply logged changes to the data in a roll forward set to bring the data forward in time.

row

In an SQL table, a single occurrence of the object modeled by the table. For example, in the AdventureWorks sample database, the Employee table models the employees of the Adventure Works Cycles company. Each row in the table records all the information about a specific employee such as an employee identification number, job title, and the date that employee was hired.

row filter

In replication, a filter that allows you to restrict the data replicated to a Subscriber based on a WHERE clause. A row filter can be static or parameterized.

row identifier

1. A column or set of columns used to distinguish any single row from every other row in the table. 2. In a heap, a pointer to the row. The pointer is built from the file identifier (ID), page number, and number of the row on the page.

row versioning

1. With cursors, the process in which the timestamp column of a row is used to determine whether data has been modified after being read into a cursor. If the data row does not have a timestamp column, row versioning is not used. 2. With transaction isolation, when a row versioning-based isolation level is enabled, the process in which the Database Engine maintains versions of data rows affected by data manipulation language (DML) execution. By choosing the appropriate isolation level, applications use row versions to query and retrieve a transactionally consistent snapshot of the data as it existed at the start of the transaction or query without acquiring locks on the data.

rowset

The OLE DB object that is used to contain a result set. It also exhibits cursor behavior depending on the rowset properties set by an application.

rule firing

The process of executing one of the application rules (event chronicle rules, subscription event rules, and subscription scheduled rules) defined in the application definition file. The rules are Transact-SQL statements that process notification data.

savepoint

A marker that allows an application to roll back part of a transaction if a minor error is encountered. The application must still commit or roll back the full transaction when it is complete.

scalar

A single-value field, as opposed to an aggregate.

scheduled backup

An automatic backup performed by SQL Server Agent when defined and scheduled as a job.

schema snapshot

In replication, a snapshot that includes the schema for published objects as well as objects required by replication (systems tables, triggers, and so on), but not user data.

scroll

The ability to move around a cursor in directions other than forward-only. Users can move up and down the cursor.

secondary database

In log shipping, a copy of the primary database that was created by restoring a database backup, with NORECOVERY or with STANDBY, on a separate server instance (the secondary server). Log backups from the primary database are restored at set intervals onto the secondary database.

securable

An object that can have a security descriptor.

security extension

A component in Reporting Services that authenticates a user or group to a report server. The default security extension in Reporting Services is Windows authentication. Custom extensions can be created to support forms-based authentication or to integrate with third-party single sign-on technologies.

segmentation

A data mining technique that analyzes data to discover mutually exclusive collections of records that share similar attributes sets. A segmentation algorithm can use unsupervised learning techniques, such as clustering, or supervised learning for a specific prediction field.

select list

The SELECT statement clause that defines the columns of the result set returned by the statement. The select list is a comma-separated list of expressions, such as column names, functions, or constants.

self-join

A join in which records from a table are combined with other records from the same table when there are matching values in the joined fields. A self-join can be an inner join or an outer join. In database diagrams, a self-join is called a reflexive relationship.

semiadditive measure

A measure that can be summed along one or more, but not all, dimensions in a cube. For example, a quantity-on-hand measure of inventory can be summed along the geography dimension to produce a total quantity on hand for all warehouses, but it cannot be summed along the time dimension because the measure specifies snapshot quantities periodically in time.

sensitive cursor

A cursor that can reflect data modifications made to underlying data by other users while the cursor is open. Updates, deletes, and inserts made by other users are reflected in the sensitive cursor. Sensitive cursors are typically used in Transact-SQL batches, stored procedures, and triggers by omitting the INSENSITIVE keyword on the DECLARE CURSOR statement.

sequenced relationship

A relationship in a repository that specifies explicit positions for each destination object within the collection of destination objects.

server cursor

A cursor implemented on the server. The cursor itself is built at the server, and only the rows fetched by an application are sent to the client.

server role

In Analysis Services, a role used to grant server-wide security privileges to a user.

Service Broker

Provides facilities for storing message queues in a SQL Server database.

set

An ordered collection of tuples with the same dimensionality.

severity level

A number that indicates the relative significance of an error that is generated by the SQL Server Database Engine. Values range from informational (1) to severe (25).

shared lock

A lock created by nonupdate (read) operations. Other users can read the data concurrently, but no transaction can acquire an exclusive lock on the data until all the shared locks have been released.

sibling

A member in a dimension hierarchy that is a child of the same parent as a specified member. For example, in a Time dimension with Year and Month levels, the members January 1997 and February 1997 are siblings.

single-user mode

A state in which only one user can access a resource. Both SQL Server instances and individual databases can be put into single-user mode.

smart tag

A smart tag exposes key configurations directly on the design surface to enhance overall design-time productivity in Visual Studio.

smart tags anchor

A smart tag anchor is a button that is displayed at the top-right edge of a component on the Windows Forms design surface. The anchor, when clicked, opens the smart tag panel.

SMTP

See Other Term: Simple Mail Transfer Protocol

Snapshot Agent

Prepares snapshot files containing schema and data of published tables, stores the files in the snapshot folder, and inserts synchronization jobs in the publication database.

snapshot replication

Snapshot replication distributes data exactly as it appears at a specific moment in time and does not monitor for updates to the data. When synchronization occurs, the entire snapshot is generated and sent to Subscribers.

snowflake schema

An extension of a star schema such that one or more dimensions are defined by multiple tables. In a snowflake schema, only primary dimension tables are joined to the fact table. Additional dimension tables are joined to primary dimension tables.

sort order

The set of rules in a collation that define how characters are evaluated in comparison operations, and the sequence in which they are sorted.

source and target

A browsing technique in which a source object is used to retrieve its target object or objects through their relationship.

source database

In data warehousing, the database from which data is extracted for use in the data warehouse.

source partition

An Analysis Services partition that is merged into another and is deleted automatically at the end of the merger process.

SQL

See Other Term: Structured Query Language (SQL)

SQL database

A database based on Structured Query Language (SQL).

SQL Mail

A component of SQL Server that allows SQL Server to send and receive mail messages through the built-in Windows NT or Windows 2000 Messaging Application Programming Interface (MAPI). A mail message can consist of short text strings, the output from a query, or an attached file.

SQL Server Authentication

One of two mechanisms for validating attempts to connect to instances of SQL Server. Users must specify a SQL Server login ID and password when they connect. The SQL Server instance verifies that the login ID and password combination are valid before allowing the connection to succeed.

SQL Server event provider (Notification Services)

A stored procedure-based event provider that gathers data from SQL Server and submits it to the event table by using the standard Notification Services event collection stored procedures.

SQL-92

The version of the SQL standard published in 1992. The international standard is ISO/IEC 9075:1992 Database Language SQL. The American National Standards Institute (ANSI) also published a corresponding standard (Data Language SQL X3.135-1192), so SQL-92 is sometimes referred to as ANSI SQL in the United States.

standby file

In a restore operation, a file that is used during the undo phase to hold a "copy-on-write" pre-image of pages that are to be modified. The standby file allows for the undo pass to be reverted so that the uncommitted transactions can be brought back. In SQL Server 2000, this is called an undo file.

star join

1. A join between a fact table (typically a large fact table) and at least two dimension tables. The fact table is joined with each dimension table on a dimension key. SQL Server considers special index manipulation strategies on these queries to minimize access to the fact table. 2. An example of a schema participating in a star join query could be a sales table, the fact table (containing millions of rows), a product table, (containing the description of several hundred products), and a store table (containing several dozen store names). In this example, the product and store tables are dimension tables. A query for selecting sales data for a small set of stores and a subset of products restricted by attributes not present in the sales database is an ideal candidate for the star query optimization.

static cursor

A cursor that shows the result set exactly as it was at the time the cursor was opened. Static cursors do not reflect updates, deletes, or inserts that are made to underlying data while the cursor is open. They are sometimes referred to as snapshot cursors.

stemmer

In Full-Text Search, for a given language, a stemmer generates inflectional forms of a particular word based on the rules of that language. Stemmers are language specific.

string

A set of contiguous bytes that contain a single character-based or binary data value. In character strings, each byte or pair of bytes represents a single alphabetic letter, special character, or number. In binary strings, the entire value is considered to be a single stream of bits that do not have any inherent pattern. For example, the constant 'I am 32.' is an 8-byte character string, while the constant 0x0205efa3 is a 4-byte binary string.

striped media set

A media set that uses multiple devices, among which each backup is distributed.

subcube

A subset of a cube that represents a filtered view of the cube, with the (All) displaying the visual totals of the members in the subcube.

Subscriber

A database instance that receives replicated data. A Subscriber can receive data from multiple Publishers and publications. Depending on the type of replication chosen, the Subscriber can also pass data changes back to the Publisher or republish the data to other Subscribers.

subscription

A request for a copy of a publication to be delivered to a Subscriber. The subscription defines what publication will be received, where, and when. There are two types of subscriptions: push and pull.

subscription event rule

A rule that processes information for event-driven subscriptions. It might contain notification generation rules or chronicle update rules. This type of rule is fired each time a corresponding batch of events is submitted.

subscription management application (Notification Services)

The application that Subscribers use to submit and manage their subscriptions. For each instance or application, a developer can create a custom interface for managing subscriptions. This application must use Notification Services APIs to submit data to the system.

subset

A selection of tables and the relationship lines between them that is part of a larger database diagram. This selection can be copied to a new database diagram. This is called subsetting the diagram.

synchronization

1. In replication, the process of data and schema changes being propagated between the Publisher and Subscribers after the initial snapshot has been applied at the Subscriber. 2. In database mirroring, when a mirroring session starts or resumes, the process in which log records of the principal database that have accumulated on the principal server are sent to the mirror server, which writes these log records to disk as quickly as possible to catch up with the principal server.

system functions

A set of built-in functions that perform operations on and return the information about values, objects, and settings in SQL Server.

system tables

Built-in tables that form the system catalog for SQL Server. System tables store all the metadata for an instance of SQL Server, including configuration information and definitions of all the databases and database objects in the instance. Users should not directly modify any system table.

table data region

A report item on a report layout that displays data in a columnar format.

table scan

A data retrieval operation where the Database Engine must read all the pages in a table to find the rows that qualify for a query.

tabular data stream (TDS)

The SQL Server internal client/server data transfer protocol. TDS allows client and server products to communicate regardless of operating-system platform, server release, or network transport.

tape backup

A SQL Server backup operation that writes to any tape device supported by Windows XP or Windows Server 2003. The tape device must be physically attached to the computer that is running the instance of SQL Server.

target server

A server that receives jobs from a master server.

tempdb database

The database that provides a storage area for temporary tables, temporary stored procedures, and other temporary working storage needs.

temporary table

A table placed in the temporary database, tempdb, and erased at the end of the session.

time dimension

A dimension that breaks time down into levels such as Year, Quarter, Month, and Day. In Analysis Services, a special type of dimension created from a date/time column.

tokenization

In text mining or Full-Text Search, the process of identifying meaningful units within strings, either at word boundaries, morphemes, or stems, so that related tokens can be grouped. For example, although San Francisco is two words, it could be treated as a single token.

trace file

A file used by SQL Server Profiler to record monitored events.

training data set

A set of known and predictable data used to train a data mining model.

trait phrasing

A way of expressing a relationship in which a minor entity describes a major entity.

transaction log

A database file in which all changes to the database are recorded. SQL Server uses transaction logs during recovery.

transaction retention period

See Other Term: distribution retention period

transactional replication

A type of replication that typically starts with a snapshot of the publication database objects and data. As soon as the initial snapshot is taken, subsequent data changes and schema modifications made at the Publisher are usually delivered to the Subscriber as they occur (in near real time). The data changes are applied to the Subscriber in the same order and within the same transaction boundaries as they occurred at the Publisher; therefore, within a publication, transactional consistency is guaranteed.

Transact-SQL cursor

A server cursor defined by using the Transact-SQL DECLARE CURSOR syntax. Transact-SQL cursors are intended for use in Transact-SQL batches, stored procedures, and triggers.

trigger

A stored procedure that executes in response to a data manipulation language (DML) or data definition language (DDL) event.

TSX

See Other Term: target server

two-phase commit

A process that ensures transactions that apply to more than one server are completed on all servers or on none.

underlying table

A table referenced by a view, cursor, or stored procedure.

unenforced relationship

A link between tables that references the primary key in one table to a foreign key in another table, and which does not check the referential integrity during INSERT and UPDATE transactions. An unenforced relationship is represented in a database diagram by a dashed line.

Unicode collation

This acts as a sort order for Unicode data. It is a set of rules that determines how SQL Server compares, collates, and presents Unicode data in response to database queries.

Union query

A query that combines two tables by performing the equivalent of appending one table onto the other.

unique index

An index in which no two rows are permitted to have the same index value, thus prohibiting duplicate index or key values. The system checks for duplicate key values when the index is created and checks each time data is added with an INSERT or UPDATE statement.

unpivot

In Integration Services, the process of creating a more normalized dataset by expanding data columns in a single record into multiple records.

update lock

A lock placed on resources (such as row, page, table) that can be updated. Updated locks are used to prevent a common form of deadlock that occurs when multiple sessions are locking resources and are potentially updating them later.

update statistics

A process that recalculates information about the distribution of key values in specified indexes. These statistics are used by the query optimizer to determine the most efficient way to execute a query.

user database

A database created by a SQL Server user and used to store application data. Most users connecting to instances of SQL Server reference user databases only, not system databases.

user-defined aggregate (function)

An aggregate function that is created by referencing a SQL Server assembly. The implementation of the user-defined aggregate function is defined in an assembly that is created in the .NET Framework common language runtime (CLR).

user-defined function

1. In SQL Server, a Transact-SQL function defined by a user. Functions encapsulate frequently performed logic in a named entity that can be called by Transact-SQL statements instead of recoding the logic in each statement. 2. In Analysis Services, a function defined in a Microsoft ActiveX library created using a Component Object Model (COM) automation language such as Visual Basic or Visual C++. Such libraries can be registered with Analysis Services and their functions called from Multidimensional Expressions (MDX) queries.

utility

A SQL Server application run from a command prompt to perform common tasks.

vacuuming (Notification Services)

The process of clearing obsolete data from a Notification Services application according to a schedule defined in the application definition file.

validation (syntactic)

The process of confirming that an XML file conforms to its schema.

variable

1. In Integration Services, stores values that can be used in scripts, expressions, and property expressions to set column values and the properties of package objects. 2. Defined entities that are assigned values. A local variable is defined with a DECLARE@localvariable statement and assigned an initial value within the statement batch where it is declared with either a SELECT or SET@localvariable statement.

vertical partitioning

To segment a single table into multiple tables based on selected columns. Each of the multiple tables has the same number of rows but fewer columns.

visual total

A displayed, aggregated cell value for a dimension member that is consistent with the displayed cell values for its displayed children. The visual total of a cell can vary from the actual total if some children of the cell are hidden. For example, if the aggregate function is SUM, the displayed cell value for Spain is 1000, and the displayed cell value for Portugal is 2000, the visual total for Iberia is 3000.

Web synchronization

In merge replication, a feature that lets you replicate data by using the HTTPS protocol.

wildcard search

The use of placeholders (such as * or ?) to search for data in a table or field. For example, you might use the string Smith* to search the Last Name field in a database for all records in which the last name starts with Smith, including Smith, Smithson, Smithlin, and so on.

Windows Management Instrumentation

An interface that provides information about objects in a managed environment. The Windows Management Instrumentation extensions for SQL are called WMI Query Language (WQL).

WMI

See Other Term: Windows Management Instrumentation

word breaker

In Full-Text Search, for a given language, a word breaker tokenizes text based on the lexical rules of the language.

word-breaking

In text mining or Full-Text Search, the process of separating strings at word boundaries. The definition of a word boundary differs greatly among natural languages and poses significant problems for natural language information retrieval.

write back

To update a cube cell value, member, or member property value.

write-ahead log

A transaction logging method in which the log is always written prior to the data.

XML Schema definition language (XSD)

A schema language. An XML Schema defines the elements, attributes, and data types that conform to the World Wide Web Consortium (W3C) XML Schema Part 1: Structures Recommendation for the XML Schema Definition Language. The W3C XML Schema Part 2: Datatypes Recommendation is the recommendation for defining data types used in XML schemas. The XML Schema definition language enables you to define the structure and data types for XML messages.

XSL

See Other Term: Extensible Stylesheet Language (XSL)

zombie

A message that was not processed by an orchestration. It is not in the Suspended queue or anywhere else where it will be retried.