BW常用术语

A

Aggregate

An aggregate is a subset of an InfoCube. The objective when using aggregates is to reduce I/O volume.The BW OLAP processor selects an appropriate aggregate during a query run or a navigation step. If no appropriate aggregate exists, the BW OLAP processor retrieves data from the original InfoCube instead.

For examples, assume you may have 10 dimensions and each one have 10 value, that means you may have up to 10^10 possible records. If high level manager often access the report focus revenue in 2008 by customer, that's mean only 2 dimension are involved. You can do aggregate and it may only contain 2 dimensions with 100 records. It reduce the I/O and improve the performance.

Aggregate rollup

Aggregate rollup is a procedure to update aggregates with new data loads. Reference Aggregate

Application component

Application components are used to organize InfoSources. They are similar to the InfoAreas used with InfoCubes. The maximum number of characters allowed for the technical name is 32.

Authorization

An authorization defines what a user can do, and to which SAP objects. For example, a user with an authorization can display and execute, but not change, a query. Authorizations are defined using authorization objects.

Authorization object

An authorization object is used to define user authorizations. It has fields with values to specify authorized activities, such as display and execution, on authorized business objects, such as queries. The maximum number of characters allowed for the technical name is 10.

Authorization profile

An authorization profile is made up of multiple authorizations. The maximum number of characters allowed for the technical name is 10.

B

BEx

Short for Business Explorer. It includes following tools to present the reports to end user: Analyzer  / Web Application Designer / Report Designer / Web Analyzer.

Bitmap index

A bitmap index uses maps of bits to locate records in a table. Bitmap indices are very effective for Boolean operations of the WHERE clause of a SELECT statement. When the cardinality of a column is low, a bitmap index size will be small, thereby reducing I/O volume.

Another possible choice in BW on this topic is B-tree index.

Business Content

Business Content is a complete set of BW objects developed by SAP to support the OLAP tasks. It contains roles, workbooks, queries, InfoCubes, key figures, characteristics, update rules, InfoSources, and extractors for SAP R/3, and other mySAP solutions.

You can simply use business content directly, or take it as template for your project to speed up implementations.

BW

BW is a Data Warehousing solution from SAP.

When we talking about BW, it often talking about BW3.5. The latest version is SAP Business Intelligence 7.0.

BW Monitor

BW Monitor displays data loading status and provides assistance in troubleshooting if errors occur. T-code: RSMO

BW Scheduler

BW Scheduler specifies when to load data. It is based on the same techniques used for scheduling R/3 background jobs.

BW Statistics

BW Statistics is a tool for recording and reporting system activity and performance information.

BWA / BIA

BW/BI Accelarator is a tool that improves the performance of BW 7.x queries on InfoCubes. It enables quick access to any data with a low amount of administrative effort and is especially useful for sophisticated scenarios with unpredictable query types, high data volume and high frequency of queries

C

Cache / OLAP Cache

A technology to improve the performance. Cache buffers query result data, in order to provider them for further accesses.  

Cache Mode

Determines whether and in what way the  query result and navigational states are to be saved in the OLAP  Cache.

Change run

Change run is a procedure used to activate characteristic data changes.

Characteristic

Characteristics are descriptions of fields, such as Customer ID, Material Number, Sales Representative ID, Unit of Measure, and Transaction Date. The maximum number of characters allowed for the technical name is 9.

Client

A client is a subset of data in an SAP system. Data shared by all clients is called client-independent data, as compared with client-dependent data. When logging on to an SAP system, a user must specify which client to use. Once in the system, the user has access to both client-dependent data and client-independent data.

Communication structure

The communication structure is the structure underlying the InfoSource.

Compound attribute

A compound attribute differentiates a characteristic to make the characteristic uniquely identifiable. For example, if the same characteristic data from different source systems mean different things, then we can add the compound attribute 0SOURSYSTEM (source system ID) to the characteristic; 0SOURSYSTEM is provided with the Business Content.

Compress

Compressing will bringing data from different requests together into one single request (request ID 0) in Infocube.

When you load data into Infocube, they are group with Requests ID. You can easily manage the data with Request ID. But the request ID have disadvantages cause the same data record (with same characteristics but different request ID) to appear more than once in the fact table. It waste the table space and affects query performance.

You can eliminate these disadvantages by compressing data and bringing data from different requests together into one single request (request ID 0).

Technically, Compress will move the data in F table to E table and aggregate the records with same char. automatically.

It is widly used when you are sure that the data in InfoCube is correct and you won't need Request ID in the future. And be careful if you defined your own delta mechanism.

D

Data packet size

For the same amount of data, the data packet size determines how work processes will be used in data loading. The smaller the data packet size, the more work processes needed.You can set it in SPRO

Data Mining

Data mining is the process of extracting hidden patterns from data. It is one of the function provided by SAP BI.

Take the most famous story "beer and diapers" as example. Wall-Mart noticed that men often bought beer at the same time they bought diapers. They mined its receipts and proved the observations. So they put diapers next to the beer coolers, and sales skyrocketed. The story is a myth, but it shows how data mining seeks to understand the relationship between different actions.

Data Warehouse

Data Warehouse is a dedicated reporting and analysis environment based on the star schema (Extended) database design technique and requiring special attention to the data ETTL process.

DataMart

The distribution of contents of ODS or InfoCube into other BW data targets on the same or on other BW systems

DataSource

A DataSource is not only a structure in which source system fields are logically grouped together, but also an object that contains ETTL-related information. Four types of DataSources exist:

• DataSources for transaction data
• DataSources for characteristic attributes
• DataSources for characteristic texts
• DataSources for characteristic hierarchies

If the source system is R/3, replicating DataSources from a source system will create identical DataSource structures in the BW system. The maximum number of characters allowed for a DataSource's technical name is 32.

DB Connect

A BW function that offers flexible options for extracting data directly into the BW from tables and views in database management systems that are connected to the BW in addition to the default connection. You can use tables and views from the database management systems that are supported by SAP to transfer data. DataSources are used to make data known to the BW, where it is then processed in the same way as data from all other sources.

Delta Caching

Additional requests will be read from the InfoCube's (F) fact table. These additional requests will be updated into the OLAP Cache. In this way, when they are changes to the data basis, the complete data for the query no longer has to be read from the database again.  

Delta update

The Delta update option in the InfoPackage definition requests BW to load only the data that have been accumulated since the last update. Before a delta update occurs, the delta process must be initialized.

Development class

A development class is a group of objects that are logically related.

Dimension table

Part of the Star Schema structure for InfoCubes. Dim tables contain pointers to the Fact tables & to the Master Data (SID tables).

Display attribute

A display attribute provides supplemental information to a characteristic. Reference Navigation attribute

Drill-down

Drill-down is a user navigation step intended to get further detailed information. For example, you can drill-down to detail report through free characteristic

DSO

Short for Data Store Object. Reference ODS in BW3.5

A DataStore object serves as a storage location for consolidated and cleansed transaction data or master data on a document (atomic) level. This data can be evaluated using a BEx query.
A DataStore object contains key fields (such as document number, document item) and data fields that, in addition to key figures, can also contain character fields (such as order status, customer). The data from a DataStore object can be updated with a delta update into InfoCubes (standard) and/or other DataStore objects or master data tables (attributes or texts) in the same system or across different systems.

There are 3 kinds of DSO: Standard DataStore Object / Write-Optimized DataStore Objects / DataStore Objects for Direct Update. Check the definition and example at:http://help.sap.com/saphelp_nw2004s/helpdata/en/f9/45503c242b4a67e10000000a114084/frameset.htm

DTP

The Data Transfer Process (DTP) transfers data from source objects to target objects in BI 7.x. You can also use the data transfer process to access InfoProvider data directly. Prerequisite: a transformation to define the data flow between the source and target object.

E 

EDW

The Enterprise Data Warehouse, a comprehensive / harmonized data warehouse solution, is design to avoid isolated applications. Reference BW330.

ETTL

ETTL, one of the most challenging tasks in building a data warehouse, is the process of extracting, transforming, transferring, and loading data correctly and quickly. Also known as ETL.

Extraction Structure

The data structure used by the extraction program to extract data. Its fields are mapped to the fields of the SAP Source System DataSource.

F

Fact tables

The Fact table is the central table of the InfoCube. Here key figures & pointers to the dimension tables are stored. There are two fact tables: the F-fact table and the E-fact table. If you upload data into an InfoCube, it is always written into the F-fact table. If you compress the data, the data is shifted from the F-fact table to the E-fact table. The F-fact tables for aggregates are always empty, since aggregates are compressed automatically. After a changerun, the F-fact table can have entries as well as when you use the functionality 'do not compress requests' for Aggregates.
E-fact table is optimized for Reading => good for Queries. F-fact table is optimized for Writing => good for Loads

Free characteristic

A free characteristic is a characteristic in a query used for drill-downs. It is not displayed in the initial result of a query run.

Full update

The Full update option in the InfoPackage definition requests BW to load all data that meet the selection criteria specified via the Select data tab. Reference Delta update

Filter

Filter is used to restrict data to a certain business sector, product group, or time period. And these can be saved and reused in the in othercontexts. Here the data is filtered and showed in the intial result of the query. 

G

Generic data extraction

Generic data extraction is a function in Business Content that allows us to create DataSources based on database views or InfoSet queries. InfoSet is similar to a view but allows outer joins between tables.

Granularity

Granularity describes the level of detail in a data warehouse. It is determined by business requirements and technology capabilities. For high level reports, manager tend to high aggregated report such as by month or by BU. But operational reports always require more detail information such as by day or by product spec.

H

High Cardinality

means that this dimension contains a high number of attributes. This information is used to execute physical optimizations, depending on the database platform. For example, different index types from those in a standard case are used. Generally, a dimension has a high cardinality if the number of dimension entries is 20% (or more) of the number of fact table entries.

I

IDoc

IDoc (Intermediate Document) is used in SAP to transfer data between two systems. It is a specific instance of a data structure called the IDoc Type, whose processing logic is defined in the IDoc Interface. See:IDOC Basics

Index

An index is a technique used to locate needed records in a database table quickly. BW uses two types of indices: B-tree indices for regular database tables and bitmap indices for fact tables and aggregate tables.

InfoArea

InfoAreas are used to organize InfoCubes and InfoObjects. Each InfoCube is assigned to an InfoArea. Through an InfoObject Catalog, each InfoObject is assigned to an InfoArea as well. The maximum number of characters allowed for the technical name is 30.

InfoCube

An InfoCube is a fact table and its associated dimension tables in the star schema. The maximum number of characters allowed for the technical name is 30.

InfoCube compression

InfoCube compression is a procedure used to aggregate multiple data loads at the request level.

InfoObject

In BW, key figures and characteristics are collectively called InfoObjects. See:https://wiki.sdn.sap.com/wiki/display/BI/Characterstic+and+Key+FiguresCharacteristic and key figures

InfoObject Catalog

InfoObject Catalogs organize InfoObjects. Two types of InfoObject Catalogs exist: one for characteristics, and one for key figures. The maximum number of characters allowed for the technical name is 30.

InfoPackage

An InfoPackage specifies when and how to load data from a given source system. BW generates a 30-digit code starting with ZPAK as an InfoPackage's technical name.

InfoSource

An InfoSource is a structure in which InfoObjects are logically grouped together. InfoCubes and characteristics interact with InfoSources to get source system data. The maximum number of characters allowed for the technical name is 32. In BW 3.5 it is a MUST and in BI 7.0 it is optional.

The difference:http://help.sap.com/saphelp_nw2004s/helpdata/en/a4/1be541f321c717e10000000a155106/content.htm

And the Recommendations for Using InfoSources in BI 7.0:http://help.sap.com/saphelp_nw2004s/helpdata/en/44/0243dd8ae1603ae10000000a1553f6/content.htm

Information Lifecycle Mangement

Generally, the data was generated in OLTP system such as R/3. In BW, Information Lifecycle Management includes data target management and data archiving.

Reference ADK and archiving for more information. And online help:http://help.sap.com/saphelp_nw2004s/helpdata/en/d0/84e5414f070640e10000000a1550b0/frameset.htm

K

Key figure

Key figures are numeric values or quantities, such as Per Unit Sales Price, Quantity Sold, and Sales Revenue. The maximum number of characters allowed for the technical name is 9. Reference Characteristic.

L

Line item dimension

A line item dimension in a fact table connects directly with the SID table of its sole characteristic.

When you creating a InfoCube, you have chance to use this feature on define dimension. It is only possible if exactly one characteristic in a dimension. It help to improve the performance because we do not need to join SID table when accessing the master data.

Logical system

A logical system is the name of a client in an SAP system.

M

Metadata Repository  

Metadata repository contains information about the metadata objects of SAP NetWeaver Business Intelligence, important object properties and their relationships with other objects. Metadata contains data about data.

Multi-cube

A multi-cube is a union of basic cubes. The multi-cube itself does not contain any data; rather, data reside in the basic cubes. To a user, the multi-cube is similar to a basic cube. When creating a query, the user can select characteristics and key figures from different basic cubes.

N

Navigational attribute

A navigational attribute indicates a characteristic-to-characteristic relationship between two characteristics. It provides supplemental information about a characteristic and enables navigation from characteristic to characteristic during a query.

The naming convention for a navigational attribute is: InfoObject_Attribute, such as 0customer_0country

Number range

A number range is a range of numbers that resides in application server memory for quick number assignments.

O

ODS

ODS is a BW architectural component located between PSA and InfoCubes that allows BEx reporting. It is not based on the star schema and is used primarily for detail reporting, rather than for dimensional analysis. ODS objects do not aggregate data as InfoCubes do. Instead, data are loaded into an ODS object by inserting new records, updating existing records, or deleting old records, as specified by the 0RECORDMODE value.

Reference DSO.

Open Hub Service

The open hub service enables you to distribute data from an SAP BW system into external data marts, analytical applications, and other applications.

OLTP / OLAP

Short for Online Transaction Processing / Online Analytical Processing.

OLTP system are Transacton-Orientated, which means it is generally used to support the daily operation. For example, SAP R/3, ERP / PDM / MES ...

OLAP is generally used to analysis data. For examples, SAP BW, DDS - Decision Support System, EIS - Executive information systems.

P

Parallel query

A parallel query uses multiple database processes, when available, to execute a query.

Partition

A partition is a piece of physical storage for database tables and indices. If the needed data reside in one or a few partitions, then only those partitions will be selected and examined by a SQL statement, thereby significantly reducing I/O volume. It can be done in Cube or DB level.

The benefit of using partition is mainly focus on:

• Well balanced partition / fragments
• Improve read performance
• Fast of delete partition by "Drop" instead of "Delete from where"

Fact table can be partitioned over a time characteristic either 0CALMONTH or 0FISCPER(Fiscal year/period). Repartition is possible in SAP BI 7.0. Seehttps://wiki.sdn.sap.com/wiki/x/N4C4AQRepartitioning

Partitioning Type

For MultiProvider queries, Partitioning Type provides options for handling the data that is returned by the individual InfoProviders from the MultiProvider separately and saving it in the cache separately.

• No Partitioning: All results of the InfoProviders are summarized as one result and cached.
• Partitioning in Groups: Infocubes in one group, and other infoproviders are saved together in groups according to their properties.
• Partitioning in Groups: Basis Cubes Separate: similar as above. individual Infocubes are saved separately.
• Each InfoProvider Separate: All InfoProviders are saved separately.

Process Chain

A process chain is a sequence of processes that are scheduled to wait in the background for an event. You can include process chains in other process chains to manage a group of them, known as meta. chains.

It is design at the implementation phase and one of the most important jobs in support and maintenance phase.

More information please reference:http://help.sap.com/saphelp_nw2004s/helpdata/en/8f/c08b3baaa59649e10000000a11402f/frameset.htm

Profile Generator

Profile Generator is a tool used to create authorization profiles.

PSA

The Persistent Staging Area (PSA) is a data staging area in BW. It allows us to check data in an intermediate location, before the data are sent to its destinations in BW.

Q

Query

A BW query is a selection of characteristics and key figures for the analysis of the data in an InfoCube. A query refers to only one InfoCube, and its result is presented in a BEx Excel sheet. The maximum number of characters allowed for the technical name is 30.

R

Read mode

Read mode determines how often the OLAP processor reads data from the database during navigation.

• Read all data
• Read during navigation
• Read during navigation & expanding hierarchy 

Reconstruct

Reconstruct is a procedure used to restore load requests from PSA.

Request

A request is a data load request from BW Scheduler. Each time that BW Scheduler loads data into an InfoCube, a unique request ID is created in the data packet dimension table of the InfoCube.

RFC

RFC (Remote Function Call) is a call to a function module in a system different from the caller's usually another SAP system on the local network.

Role

In Profile Generator, an authorization profile corresponds to a role. A user assigned to that role also has the corresponding authorization profile. A user can be assigned to multiple roles. The maximum number of characters allowed for the technical name is 30.

Reclustering

It allows you to change the clustering of Infocubes and Data store objects that already contain data.

S

SID

SID (Surrogate-ID) translates a potentially long key for an InfoObject into a short four-byte integer, which saves I/O and memory during OLAP.

Source system

A source system is a protocol that BW uses to find and extract data. When the source system is a non-SAP system, such as a flat file or a third-party tool, the maximum number of characters allowed for the technical name is 10. When the source system is an SAP system, either R/3 or BW, the technical name matches the logical system name. The maximum number of characters allowed for the technical name is 32.

Star schema

A star schema is a technique used in the data warehouse database design to help data retrieval for online analytical processing.

SAP BW use extended star schema with following advantage compare to classical star schema:

• Use of generated key (short number) for faster access
• External hierarchy
• Multi-language support
• Master data are separated. It is common to all cubes
• Support slowly changing dimensions
• Aggregates in its own tables for faster access

Statistics

For a SQL statement, many execution plans are possible. The database optimizer generates the most efficient execution plan based on either the heuristic ranking of available execution plans or the cost calculation of available execution plans. Statistics is the information that the cost-based optimizer uses to calculate the cost of available execution plans and select the most appropriate one for execution. BW uses the cost-base optimizer for Oracle databases.

System Administration Assistant

System Administration Assistant is a collection of tools used to monitor and analyze general system operation conditions.

System landscape

The system landscape specifies the role of each system and the paths used in transporting objects among the various systems.

T

Time-dependent entire hierarchy

A time-dependent entire hierarchy is a time-dependent hierarchy whose nodes and leaves are not time-dependent.

For example, you have two hierarchy which presents organization of company in 2007 and 2008.

Time-dependent hierarchy structure

A time-dependent hierarchy structure consists of nodes or leaves that are time-dependent, but the hierarchy itself is not time-dependent.

For example, you have one hierarchy but one of his node was changed from 2007 and 2008. HR may belong to administration division before 2007 and belong to supporting division after 2008.

Transfer rule

Transfer rules specify how DataSource fields are mapped to InfoSource InfoObjects.

In BI 7.0, transfer rule and update rule are replaced by transformation.

Transfer structure

A transfer structure maps DataSource fields to InfoSource InfoObjects.

Transformations (TRF)
TRFs connect source objects (DataSources, DSOs, InfoCubes, InfoSets) to Data Targets (InfoProviders) of a data-staging process. They replace combination of transfer rules & update rules. The TRF process allows you to consolidate, cleanse & integrate data. You must use DTP to load data to InfoProvider.

U V W X Y Z

Update rule

An update rule specifies how data will be updated into their targets. The data target can be an InfoCube or an ODS object. If the update rule is applied to data from an InfoSource, the update rule's technical name will match the InfoSource's technical name. If the update rule is applied to data from an ODS object, the update rule's technical name will match the ODS object's technical name prefixed with number 8.

In BI 7.0, transfer rule and update rule are replaced by transformation. But we still have chance to use it in BI 7.0

Variable

A variable is a query parameter. It gets its value from user input or takes a default value set by the variable creator.

Variables act as placeholders for:

• Characteristic values
• Hierarchies
• Hierarchy nodes