bex monitor

BEx Monitor

Release 350

AD

DO

N.N

DB

WG

EN

SAP Online Help 08.08.03

BEx Monitor 350 2

Copyright

© Copyright 2003 SAP AG. All rights reserved.

No part of this publication may be reproduced or transmitted in any form or for any purposewithout the express permission of SAP AG. The information contained herein may be changedwithout prior notice.Some software products marketed by SAP AG and its distributors contain proprietary softwarecomponents of other software vendors.

Microsoft®, WINDOWS®, NT®, EXCEL®, Word®, PowerPoint® and SQL Server® are registeredtrademarks of Microsoft Corporation.

IBM®, DB2®, DB2 Universal Database, OS/2®, Parallel Sysplex®, MVS/ESA, AIX®, S/390®,AS/400®, OS/390®, OS/400®, iSeries, pSeries, xSeries, zSeries, z/OS, AFP, Intelligent Miner,WebSphere®, Netfinity®, Tivoli®, Informix and Informix® Dynamic ServerTM are trademarks ofIBM Corporation in USA and/or other countries.

ORACLE® is a registered trademark of ORACLE Corporation.

UNIX®, X/Open®, OSF/1®, and Motif® are registered trademarks of the Open Group.

Citrix®, the Citrix logo, ICA®, Program Neighborhood®, MetaFrame®, WinFrame®,VideoFrame®, MultiWin® and other Citrix product names referenced herein are trademarks ofCitrix Systems, Inc.HTML, DHTML, XML, XHTML are trademarks or registered trademarks of W3C®, World WideWeb Consortium, Massachusetts Institute of Technology.

JAVA® is a registered trademark of Sun Microsystems, Inc.

JAVASCRIPT® is a registered trademark of Sun Microsystems, Inc., used under license fortechnology invented and implemented by Netscape.

MarketSet and Enterprise Buyer are jointly owned trademarks of SAP AG and Commerce One.SAP, SAP Logo, R/2, R/3, mySAP, mySAP.com and other SAP products and services mentionedherein as well as their respective logos are trademarks or registered trademarks of SAP AG inGermany and in several other countries all over the world. All other product and service namesmentioned are trademarks of their respective companies.


BEx Monitor 350 3

Icons

Icon Meaning

Caution

Example

Note

Recommendation

Syntax

Typographic Conventions

Type Style Description

Example text Words or characters that appear on the screen. These include fieldnames, screen titles, pushbuttons as well as menu names, paths andoptions.

Cross-references to other documentation.Example text Emphasized words or phrases in body text, titles of graphics and tables.

EXAMPLE TEXT Names of elements in the system. These include report names,program names, transaction codes, table names, and individual keywords of a programming language, when surrounded by body text, forexample, SELECT and INCLUDE.

Example text Screen output. This includes file and directory names and their paths,messages, source code, names of variables and parameters as well asnames of installation, upgrade and database tools.

EXAMPLE TEXT Keys on the keyboard, for example, function keys (such as F2) or theENTER key.

Example text Exact user entry. These are words or characters that you enter in thesystem exactly as they appear in the documentation.

<Example text> Variable user entry. Pointed brackets indicate that you replace thesewords and characters with appropriate entries.


BEx Monitor 350 4

BEx Monitor ..................................................................................................................................... 5

OLAP Trace Tool ......................................................................................................................... 6

Setting an OLAP Trace ............................................................................................................ 9

Query Monitor ............................................................................................................................ 10

Query Properties .................................................................................................................... 13

Read mode ......................................................................................................................... 15

Cache Mode........................................................................................................................ 17

Accuracy ............................................................................................................................. 20

Not Using Parallel Processing ............................................................................................ 21

Using the Cache Despite Virtual Characteristics and Key Figures .................................... 22

Technical Information ............................................................................................................. 24

OLAP: Cache Monitor ................................................................................................................ 28

Global Cache Parameters ...................................................................................................... 32

Cache Data Removal and Swapping ..................................................................................... 35

Cache Structure...................................................................................................................... 38

Logical File Name................................................................................................................... 41


BEx Monitor 350 5

BEx MonitorPurposeIn the Business Information Warehouse SAP Easy Access Menu, under Business Explorer, youcan find a range of entries that allow the administrator to check the execution of queries or queryviews in the backend of the BW system.

You can also change settings for InfoProviders on an individual query level in Customizing.

Implementation ConsiderationsThe SAP Easy Access initial screen was introduced in BW 3.0A SP11 = BW 3.0B SP04.

FeaturesQuery Monitor

For more information, see Query Monitor [Seite 10].

OLAP Trace Tool For more information, see OLAP Trace Tool [Seite 6].

OLAP: Cache Monitor For more information, see OLAP Cache Monitor [Seite 28].

ICM Monitor For more information, see Monitoring the ICM with the ICM Monitor [Extern].


BEx Monitor 350 6

OLAP Trace Tool PurposeThe OLAP Trace tool allows you to trace data transferred to the BW system as a result of actionsmade in the frontend. You can also trace values for variables. The OLAP trace is stored in theBW system.

You can use the OLAP trace to reproduce activities in the frontend. The OLAP trace tool isparticularly useful when testing and when examining frontend activities involving SAP support.

The log itself, created when an OLAP trace is created in the frontend system, is of little use.

Implementation ConsiderationsNote the following: ...

1. Keep the trace as short as possible. It may be useful to log the problem with several traces.

2. The trace is no longer valid once the query for which the trace was set is changed. Inparticular, the trace loses its validity when carrying out the following actions:

� Changing the hierarchy

� Changing restrictions

� Adding or removing characteristics and key figures

� Reading data (The trace finds nothing if the data for a navigation step that was traced isall of a sudden no longer there).

3. When using the OLAP trace tool to assist SAP Support in investigating a problem, you alsoneed to make sure that the test user you make available to SAP Support has thenecessary authorizations.

� Authorization for executing the query with all navigation steps and data requests.

� Authorization for the OLAP trace tool (RSRTRACE or RSRCATTTRACE)

IntegrationAs of BW 3.0A SP11 = BW 3.0B SP04, you can reach the OLAP trace tool from the SAP EasyAccess screen, from Business Explorer � BEx Monitor � Trace Tool.

Further information about setting an OLAP trace can be found under Setting an OLAP Trace[Seite 9].

FeaturesThe following functions are available in the OLAP trace tool toolbar:

Activate User

The user from the User field is added to the list. This user’s actions are logged.

All users

The system puts the entry ALLUSERS into the list. All users are now logged. Thoseusers that were already activated retain their own configuration however.


BEx Monitor 350 7

Deactivate User

The user selected in the table is deleted from the list. This user’s actions are not loggedany more.

Configure User

You get to the Set Trace Configuration screen. Enter a standard description for the activelog process. You can also choose topic areas and individual routines that you should lookat in greater detail. You can also activate the Detailed Runtime Analysis.

User Logs

The Processing/Deletion of Logged Callups screen appears. The system displays the logsthat belong to the current user.

CATT Logs

All Logs

The Processing/Deletion of Logged Callups screen appears. The system displays all OLAPtraces set in the BW system.

Navigating within LogsYou are in the Processing/Deletion of Logged Callups screen. Each entry in the list contains thefollowing entries for each log:

� System

� Log No.

� User

� Date

� Time

� Description (If necessary)

Delete the logs you no longer need. The number of logs is restricted by the system.When the maximum number is reached, the system starts to delete the oldest logs.

Double-clicking on a log takes you to the “Call up tree” Each entry in the list can contain entriesfor the following characteristics for the log considered in greater detail:

� SUB (The nesting depth is displayed with an “X” or “XX” in this column).

� Function module

� Parameter (for function module)

� InfoCube

� Report name

� CPU time (µs) (required by the trace).

You get back to the Set Trace Configuration screen via Configuration. You can entera description of the current log here.


BEx Monitor 350 8

Double-clicking on a row takes you to the ABAP Debugger. The log is executed up until thefunction module is called up (0 nesting depth). If a module with subrc <>0 that was also describedis to be exited, a breakpoint is inserted here.

The logs can also be executed via remote login.


BEx Monitor 350 9

Setting an OLAP Trace Procedure ...

1. Activate the trace mode.

You have various options to activate the trace mode for all additional actions on thefrontend:

� Call up the BEx Analyzer. Via the Settings, choose Trace, before logging into the BWsystem.

� From the Business Explorer SAP Easy Access screen, choose Business Explorer� BExMonitor �Trace Tool. Enter a user and choose Edit � Activate user. You can activate allusers via All Users. For more information, see OLAP Trace Tool [Seite 6].

To activate the trace mode for a particular query in the Web, you need to add theparameter &trace=X to the query URL.

2. Log into the BW system and carry out the required steps.

The frontend actions are logged. The log file SAPBEX.trc is generated on the frontendcomputer.

Make sure that you keep the log as short as possible.

To restrict logging to important steps, it is useful to start the BEx Analyzer without thetrace at first, then to load the workbook. Only at this point should you interrupt theconnection to the OLAP server, via Settings, and activate the trace. You can thencreate a new connection via Refresh. The actions you subsequently carry out arenow logged.

3. Deactivate the trace mode in the same way you activated it in step 1.

ResultYou have logged actions in the frontend. The logs and the OLAP trace number can be found inthe OLAP Trace Tool.

When making contact with SAP support, your customer message can contain the technical name(from the InfoCube or query) as well as the number of the OLAP Trace (Log Number) in additionto the user, and the date and time the log was created.


BEx Monitor 350 10

Query MonitorPurposeThe query monitor tests, checks, and manages BW queries. By using the query monitor, you cantest queries or query views, as well as regenerate, check, or change query properties.

With the query monitor you have the option of displaying technical information about queries.Furthermore, the query monitor allows the entry into the cache monitor.

The query monitor is a transaction for administrators, because previous technical knowledge ispresumed.

Implementation NotesYou can use this component from the BW system without a BW Frontend. However, aprerequisite would be that you have already defined the BW queries you want in the BEx QueryDesigner.

IntegrationAs of BW 3.0A SP11 = BW 3.0B SP04, you can reach the query monitor from SAP Easy Accessscreen from Business Explorer � BEx Monitor � Query Monitor.

FeaturesQuery

Specify the technical name for the BW query you want in the entry field. Input help is available forsearching in History, InfoAreas, Roles, and Favorites.

View

If you want to test a query view, specify the technical name for the query view you want in theentry field. Input help is available.

The following functions are available in the query monitor toolbar:

Pushbutton Function

Execute Execute the query or query view

You arrive in a screen with the query display in an HTML control. Youcan test the functions with certain restrictions in the display as youwould on the Web.

In the upper part of the entry screen, the URL is displayed with thecomplete parameterization.

By using the respective pushbuttons, you can set bookmarks

Generate Report Regenerates the query code. The code is stored as a generatedreport.

With a regeneration, the agreement is always given with the querydefinition


BEx Monitor 350 11

Properties You arrive at the query properties dialog box.

You can find additional information under Query Properties [Seite 13].

Messages You arrive at the Suppress Messages dialog box.

In this screen you can determine which messages for the chosenquery are not to be shown to the user in the Frontend.

In the left-hand screen area you are able to choose the messagecategory, for example, Generating, Hierarchy, Variables, Key Figures,Calculate as. In the right-hand screen area you see those messagesyou can choose not to display. If you double-click on the short text forthe message the system displays the long text for the message in thescreen area below.

Set the indicator in the ready for impute field for the message that youwant to suppress, and go back. The system saves your entriesautomatically

By using the respective pushbuttons, the following functions are available for selection:

Pushbutton Function

Performance Info You get to the Performance Info screen. The system displaysperformance-relevant information for the query that do notcorrespond to the system recommendations ( ). The informationrefers to the following areas:

Query definition:

� Query cannot use aggregates (corresponds to specifications inTechnical Information [Seite 24] under OLAP-Relevant Data)

� Read mode X or A (see Read Mode [Seite 15])

� Query cannot use the cache (corresponds to specifications inTechnical Information [Seite 24] under OLAP-relevant Data)

� Query uses customer exits

� There are non-cumulative values with AVI (corresponds tospecifications in Technical Information [Seite 24] under Non-cumulative Flags)

InfoProvider:

� InfoProvider is a MultiProvider

� Database statistics need to be checked

� Database indexes need to be checked

Technical Info You arrive at the Technical Information screen.

You can find additional information under Technical Information [Seite24].

Cache Monitor You arrive at the Cache Monitor screen.

For more information, see OLAP Cache Monitor [Seite 28].


BEx Monitor 350 12

Query Variants If the query contains variables that are ready for entry, the nextscreen displays the query variants. You can change these or createnew variants.

You can find additional information about query variants under SavingVariants for a Query [Extern].

Test IGS Test Internet Graphics Server

With the Internet Graphics Server (IGS) from SAP, you can createcharts and maps in the BW system for Web-based reporting.

By using Test IGS, the Internet Graphics Server is checked to seewhether it is accessible. If the system displays a graphic with two setsof three bars, you have a connection. You can now connect toperformance analyses.

You can find additional information in the implementation guide (IMG)of the Business Information Warehouse under Reporting-RelevantSettings � Web-Based Settings � IGS Settings.

RestrictionsThe function that offers the BW Frontend is not entirely supported in the query monitor. On theother hand, there are settings that you can only make here and not in the Frontend (see Properties and Query Properties [Seite 13]).


BEx Monitor 350 13

Query PropertiesUseYou can make settings for a BW query concerning the Read Mode, the Cache Mode, and thefrom the Query Properties dialog box for the query monitor. You can switch off the default ParallelProcessing for queries on a MultiProvider. For queries with virtual characteristics or key figuresyou are able to activate the use of the cache, which is deactivated by default.

IntegrationRead Mode and Cache ModeIn Customizing, you can specify the standard settings for an InfoProvider for the read mode andthe cache mode. These need to be valid for those queries that will be created for thecorresponding InfoProvider. Furthermore, for the InfoProvider for which data changes do notoccur under the control of the BW system (for example, virtual InfoCubes and transactional ODSobjects), you can specify the cache validity, meaning the data retention period in the cache.

You can find additional information in the implementation guide (IMG) of theBusiness Information Warehouse under Reporting-Relevant Settings � GeneralReporting-Settings � InfoProvider Settings.

In the Query Monitor, you can change the standard settings for the read mode and the cachemode for the queries that already exist. You can determine the cache validity exclusively for theInfoProvider in Customizing.

AccuracyYou can specify the accuracy with which a key figure needs to be calculated in the InfoObjectmaintenance.

You can only specify the accuracy with which all query key figures need to be calculated in theQuery Monitor (and not in the BEx Query Designer from the BW Frontend). You can use this to“override” the affected settings for individual key figures in the InfoObject maintenance, ifnecessary.

Parallel ProcessingA query on a MultiProvider is deconstructed internally into sub-queries. These are processed inparallel by default. In most cases parallel processing has benefits in terms of performance.However, in some cases sequential processing can be more advantageous for queryperformance (see Parallel Processing of Sub-Queries vs. Sequential Processing [Extern]).

You are able to switch parallel processing for a MultiProvider off in the RSADMIN table.

You are able to switch parallel processing for a query on a MultiProvider off in the Query Monitor.

PrerequisitesYou have created the query you want in the BEx Query Designer. The settings specified for thebasic InfoProvider in Customizing, or those that were already specified earlier in the QueryMonitor, apply to the query (see above in the Integration section)


BEx Monitor 350 14

Features� Read Mode [Seite 15]

� Cache Mode [Seite 17]

� Accuracy [Seite 20]

� Not Using Parallel Processing [Seite 21]

� Using the Cache Despite Virtual Characteristics and Key Figures [Seite 22]


BEx Monitor 350 15

Read modeDefinitionThe read mode determines how the OLAP processor gets data during navigation. You can set themode in Customizing for an InfoProvider and in the Query Monitor for a query.

UseThe following types are supported: ...

1. Query to be read when you navigate or expand hierarchies (H)

The amount of data transferred from the database to the OLAP processor is the smallest inthis mode. However, it has the highest number of read processes.

In the following mode Query to Read Data during Navigation, the data for the fullyexpanded hierarchy is requested for a hierarchy drilldown. In the Query to be read whenyou navigate or expand hierarchies mode, the data across the hierarchy is aggregated andtransferred to the OLAP processor on the hierarchy level that is the lowest in the start list.When expanding a hierarchy node, the children of this node are then read.

You can improve the performance of queries with large presentation hierarchies bycreating aggregates on a middle hierarchy level that is greater or the same as the hierarchystart level.

2. Query to Read Data during Navigation (X)

The OLAP processor only requests data that is needed for each navigational status of thequery in the Business Explorer. The data that is needed is read for each step in thenavigation.

In contrast to the Query to be read when you navigate or expand hierarchies mode,presentation hierarchies are always imported completely on a leaf level here.

The OLAP processor can read data from the main memory when the nodes are expanded.

When accessing the database, the best aggregate table is used and, if possible, data isaggregated in the database.

3. Query to Read All Data At Once (A)

There is only one read process in this mode. When you execute the query in the BusinessExplorer, all data in the main memory area of the OLAP processor that is needed for allpossible navigational steps of this query is read. During navigation, all new navigationalstates are aggregated and calculated from the data from the main memory.

The read mode Query to be read when you navigate or expand hierarchies significantly improvesperformance in almost all cases compared to the other two modes. The reason for this is that onlythe data the user wants to see is requested in this mode.

Compared to the Query to be read when you navigate or expand hierarchies, the setting Query toRead Data During Navigation only improves performance for queries with presentationhierarchies.

Unlike the other two modes, the setting Query to Read All Data At Once also improvesperformance for queries with free characteristics. The OLAP processor aggregates on thecorresponding query view. For this reason, the aggregation concept, that is, working with pre-aggregated data, is least supported in the Query to Read All Data At Once mode.


BEx Monitor 350 16

We recommend you choose the mode Query to be read when you navigate orexpand hierarchies.

Only choose a different read mode in exceptional circumstances. The read modeQuery to Read All Data At Once may be of use in the following cases:

� The InfoProvider does not support selection. The OLAP processor readssignificantly more data than the query needs anyway.

� A user exit is active in a query. This prevents data from already beingaggregated in the database.

IntegrationYou can find additional information under Query Properties [Seite 13].


BEx Monitor 350 17

Cache ModeDefinitionThe cache mode determines whether and in what ways the query results and navigational statescalculated by the OLAP processor as highly compressed data are to be saved in a cache. Youcan set the mode in Customizing as the standard value for an InfoProvider, and in the QueryMonitor for a query.

UseCaching is a means of improving query performance. However care should be taken to balancebenefits against costs:

� Benefit: Caching improves the answer time for queries

� Costs: Expenditure involved in organizing and retaining data in the cache

You are able to hold the cache data in the main memory, distributed to an application server or ina network. The option that you choose depends on various parameters. These include:

� How often the query is requested

We recommend that you save queries that are requested particularly frequently in thecache. The main memory cache is particularly fast but restricted by its size. Swappingcached data does cancel out limitations on the main memory but simultaneously affectsperformance. There are practically no limitations on memory space available in thedatabase or in the file system for cluster and flat file caches. Accessing compressed datadirectly in cluster and flat file caches is also very beneficial in terms of performance.

� The complexity of the query

Caching does bring advantages in terms of performance, particularly with more complexqueries as evaluating these requires a higher level of effort. We recommend that complexdata processed by the OLAP processor is held in the cache. (Therefore the cache modeMain Memory Without Swapping is less suitable for queries of this type.)

� How often data is loaded

Using the cache is barely advantageous if query-relevant data is often changed andtherefore has to be loaded frequently. The cache has to be regenerated every time. Ifcached data is held in the main memory, data from frequently called-up queries can bedisplaced. Subsequently your call-up takes more time.

StructureThe following modes are supported for the global cache: ...

1. Cache is Inactive (0)

The cross-transactional cache is deactivated.

The following modes are possible for the active cache: The cross-transactional cache isactivated.

2. Main Memory Cache Without Swapping (1)

The cache data is stored in the main memory. When the cache memory has beenexhausted, excess data is removed according to the LRU algorithm, i.e. deleted from thememory. When a query is subsequently requested, the result is read from the InfoProvideronce again.


BEx Monitor 350 18

3. Main Memory Cache with Swapping (2)

The cache data is stored in the main memory. When the caching memory is used up, theexcess data is written to a background store (file or cluster-table) and can be loaded backinto the cache memory when making another request.

After query-relevant data has been loaded again, the background store can no longerbe used.

4. Cluster /Flat File Cache per Application Server (3)

The cache data is stored persistently as cluster tables in a database or as a file in adirectory attainable from the application server. We recommend that you choose adirectory that is proximate to the application server.

When you save the cache data in a cluster table this puts a strain on the database.However, it is also then quicker to access uncached data in the database. In comparisonto storing data in, for example, a star schema (see InfoCube [Extern]), fewer tables areread and fewer indexes are accessed.

5. Cross-Application Server Cluster / Flat File Cache (4)

The cache data is stored persistently as a cross-application server cluster table or file in afile system in the network, accessed from the application server. In this mode there is nodisplacement of data and no restriction on memory size. More space is required but thismethod also saves time.

� Cluster table: Strain is put on the database instance for the cluster table.

� File: Strain is put on the network node operating system for the file.

� In addition, the application server that calculates the data, and networkcommunication have to be taken into account with both storage types.

The following table offers an overview of the available modes for an active cache:

Active cache mode

Main memory cache Cluster / flat file cacheProperty

Withoutswapping

With swapping per applicationserver

Cross-applicationserver

Removal Yes (LRUalgorithm anddelete)

Yes (LRUalgorithm andswapping)

No No

Swapping No Yes No No

Global cachesize

Restricted Main memoryrestricted,outsourced part notrestricted

Unrestricted Unrestricted

File name No With swapping inflat files: The dataremoved is to befound under the filename

With flat file caches:All data is to befound under the filename Theapplication server isone element of thefile name

With flat file caches:All data is to befound under the filename Theapplication server isnot an element ofthe file name


BEx Monitor 350 19

IntegrationYou can find additional information about the various ways of setting the cache mode for anInfoProvider or a query under Query Properties [Seite 13].

You can find additional information about the global cache parameter and the amount of memoryused by the runtime object under OLAP Cache Monitor [Seite 28].


BEx Monitor 350 20

AccuracyDefinitionThe determination of a specific accuracy decides which number format is used internally forcalculating key figures. The degree of accuracy is determined in the InfoObject maintenance foreach key figure and in the query monitor for all query key figures.

UseWhen converting decimal figures with numbers following the decimal point into binary floating-point numbers, unavoidable rounding errors occur. This is because many decimals (for example,0.1) correspond to periodic fractions, and are therefore unable to be displayed exactly.

If you choose Calculate with Packed Numbers in the query monitor, packed numbers (ABAP datatype P, with nine decimal places) are used as the internal number format for all query key figuresinstead of binary decimals. Rounding errors are avoided with commercial calculations with thisfixed-point number, because in doing so no conversion of numbers into the dual system takesplace.

Choose this option for those application areas in which a high accuracy is required (for example,for queries in the accounting and cost accounting areas).

Because calculating with packed numbers causes a high processing status overall,we recommend not using this number format for all query key figures, but rather onlyfor those key figures for which a very high degree of accuracy is required.

You have to make this setting in the InfoObject maintenance for the respective keyfigure:

i. From the SAP Easy Access Menu, choose Modeling � Object Maintenance� InfoObject.

ii. Enter the name of the key figure and choose Maintain.

iii. To calculate with packed numbers, choose from the Additional Properties tabpage under Diverse with the option Key Figure with Maximum Accuracy.


BEx Monitor 350 21

Not Using Parallel ProcessingUseParallel processing of a MultiProvider query is cancelled as soon as the overall result of sub-accesses to the InfoProvider involved with the MultiProvider exceeds 30,000 rows. TheMultiProvider query is then automatically restarted and processed sequentially.

Because of this it seems as though it is quicker to process queries of this type sequentially. Inreality, the parallel processing that is cancelled is equivalent to the subsequently executedsequential processing.

To save yourself unnecessary costs incurred by canceling parallel processing, change tosequential processing for these crucial queries.

IntegrationRegarding entries in RSADMIN, follow these proposed solutions:

� Switch off parallel processing for the MultiProvider for all queries

� Copy the MultiProvider, switch off parallel processing for the copied MultiProvider, andassign queries not suitable for parallel processing to this MultiProvider

� Parameterize the maximum possible size of overall results.

You will find additional information under Parallel Processing of Sub-Queries vs. SequentialProcessing [Extern].

PrerequisitesYou have identified the crucial queries in one of the following ways:

� With MultiProviders that have the BW statistics data on queries specified, you are able toidentify those queries that will have values of 30,000 or higher in the QDBTRANS columnby using table RSDDSTAT (transaction SE16).

Make sure that not all queries executed in RSDDSTAT with QDBTRANS>= 30000are included, but just those that are defined for a MultiProvider.

� You are able to execute a query in transaction RSRT using Execute + Debug" with theMultiprovider Explain debug option. The system displays a message if parallel processingwas cancelled and sequential processing started when a query was executed.

FeaturesIf you choose the option Do Not Use Parallel Processing the selected query will be sequentiallyprocessed in future.


BEx Monitor 350 22

Using the Cache Despite Virtual Characteristics andKey FiguresUseUsing the OLAP cache serves to improve performance when executing queries (see OLAPCache Monitor [Seite 28] and Cache Mode [Seite 17]).

If virtual characteristics or key figures are used in the query, the cache is deactivated by default.These InfoObjects are user-specific in the way that they are defined and evaluated by using acustomer exit (SAP Enhancement RSR00002 or BAdI RSR_OLAP_BADI). They are only filledwith data at query runtime.

If you want to use the cache for a query that contains virtual characteristics or key figuresanyway, you can choose the option Using the Cache Despite Virtual Characteristics and KeyFigures in the Query Properties dialog box.

Make sure that the customer exit from which the virtual characteristic or key figureobtain your data during runtime is independent of the cache. Otherwiseinconsistencies can occur in the reported data.

PrerequisitesSAP Enhancement RSR00002 ...

1. You have created an active project in transaction CMOD, which you have assigned to SAPEnhancement RSR00002.

2. You have defined the virtual characteristics or key figures desired in the InfoProvider. Youhave defined customer-specific code in include reports for the function moduleEXIT_SAPMRSRU_001.

You can find additional information on customer exits in the system documentation.In transaction SMOD choose sub-object Documentation for enhancementRSR00002 (BW: Virtual Characteristics and Key Figures in Reporting). Choose Display.

BAdI RSR_OLAP_BADI

Alternatively you can implement the Business Add-In RSR_OLAP_BADI (Virtual Characteristicsand Key Figures from SMOD) instead of SAP Enhancement RSR00002.

You can find additional information on the interface of this Business Add-In in BAdIdefinition (SE19). Take care that you are able to copy example code to this BAdI(example implementation class CL_EXM_IM_RSR_OLAP_BADI).

ExampleThe following ODS objects for the utilities industry, delivered in the context of SAP BusinessContent, contain examples of virtual characteristics and key figures:


BEx Monitor 350 23

InfoProvider Virtual InfoObjects

ODS object Stock Statistics Prepayment Meter

Technical name: 0UCS_DS02

Virtual characteristics:

VKF key figure area consumption debt

VKF key figure area consumption debt rate

VKF key figure area non consumption debt

VKF key figure area non consumption debt rate

ODS object Payment Scheme

Technical name: 0UCS_DS03

Virtual key figure

Number of payments


BEx Monitor 350 24

Technical InformationUseFrom the Technical Information screen, you get an overview of important, technical, referencenumbers for a query.

IntegrationIn order to access the Technical Information screen, choose a query from the Query Monitorscreen and then Technical Info.

PrerequisitesThe technical information is automatically created.

FeaturesThe technical information is grouped by category for the selected query. Depending on the querydefinition, all or only a part of the categories is displayed. The first line of the HTML controlcontains links to the categories displayed in table form in the lower area of the page.

System-side entries are specified in the first column for some of the categories:

The symbol means that the value of the technical reference numbers correspondsexactly to the system-side recommendation.

The symbol means that the value of the technical reference numbers possibly does notcorrespond to the system-side recommendation.

The symbol means that the value of the technical reference numbers does notcorrespond to the system-side recommendation. Therefore, a change can be quitemeaningful. Such a calculation does not necessary mean that an error must exist.

General Data

� Data target(s)

Query Definition Data

� Technical name of the query definition

� Last change to the query definition (date and time)

OLAP-Relevant Data

� Query name (technical name and internal name)

� Name of the generated programs (by using Display Gen. Report, you can display theprogram; see below).

� Query generation time (see under Cache-Relevant Data)

� Version (external/internal)

� Generation status (active/inactive)

� Read mode (You can find additional information under Read Mode [Seite 15].)

� Released for OLE DB for OLAP (You can find additional information under QueryProperties [Extern].)


BEx Monitor 350 25

� Calculating before the aggregation (You can find additional information under CalculatedKey Figure Properties [Extern].)

� Virtual characteristics/key figures

� Attributes in calculated key figures

� Currency translation in the query definition

� Query can use aggregates

� Value field groups (groups of key figure IDs or selection elements and basic sizes for theform elements, combined by specific properties, for example, their aggregation behavior orby specific steps of the process that are required, for example, calculating as packednumbers)

� Key figure IDs for display (KIDs)

� Calculated key figure IDs

� Selecting Constants (You can find additional information under Selecting Constants[Extern].)

� Fiscal year variant

� Query key date

� Link to the Cache-Relevant Data table (see below)

Non-Cumulative Flags

� There is a non-cumulative value with AVI

� There are non-cumulative values with LAST or FIRST

� There are non-cumulative values with MAX or MIN

You can find additional information about the types of exception aggregation for non-cumulative values under Aggregation Behavior of Non-Cumulative Values [Extern].

� More than a provider of non-cumulatives (see Evaluating Non-Cumulative Values [Extern]in the InfoCube definition section)

� NCUMTIM is not unique (see Temporal Characteristic [Extern])

� Query does not interrupt the critical interval (see Validity Period [Extern])

� The non-cumulative method L1 is required

� The non-cumulative method L2 is required

StatisticsThis section runs the statistics by specifying the current and maximum number:

� InfoCubes involved for the MultiProvider

� FEMS number (selection groups)

� Number of elements in structure 1 and 2 (for the structures in the columns and rows)

� Selection rows

� Columns in the SP storage table

� Free characteristics

� Basic key figures


BEx Monitor 350 26

� Formula components

� Hierarchies

� Hierarchy nodes

� Variables

Invisible Drill-Down CharacteristicsIf a query has invisible drill-down characteristics, they are displayed in this section with theirtechnical name and description, for example:

0CURRENCY Units or currency field

Cache-Relevant Data

� Query can use the cache (if necessary, with specifying the reason why the query cannotuse the cache)

In addition, this section includes important time stamps:

� Query generation time (corresponds to the specification under OLAP-Relevant Data)

� Time of the last loading of data into the InfoProvider (last data change)

� Last change to data value set (pre-calculated value set)

� Last master data change for navigation attributes

This time stamp influences the invalidation of the cache. If the last time stamp for acache entry for the selected query is older than one of the above-named timestamps, then the cache is rebuilt.

Optimization Information

� Optimizes table sizes

After a query has been executed for the first time, the OLAP processor determines theoptimal initial size of internal tables and generates the report automatically afterwards. Thesystem shows the date and time of this last update. If the query definition is not changed, anew optimization does not take place automatically. You are able to force a newoptimization however by choosing the link Optimize Again.

Load Size for a Generated ReportThis section restricts ABAP programs. If these are not complied with, then the report is notgenerated, and the query is not executed.

� PXA buffer

� Static control blocks

� Variable control blocks

� Literals

� Structure descriptions

� Includes

The following functions are available in the screen toolbar:

Function What you need to know


BEx Monitor 350 27

Display Gen. Report Goes to the ABAP editor.

You see the ABAP program that was generated forthe query you selected. (You can find the name ofthe generated report under OLAP-Relevant Data)

Query Def. Designer Goes to the ABAP debugger.

You see the coding for the interface between theBEx Query Designer and the OLAP processor

Query Def. Processor Goes to the ABAP debugger.

You see a few lines of the coding according to theabove-named interface. For this break point, thecomplex data structure SX_REPORT and internalcontrol tables are displayed for the query


BEx Monitor 350 28

OLAP: Cache MonitorPurposeFrom the Cache Monitor screen, you can look at the global cache parameters, the amount ofmemory used by the query runtime objects and the current underlying cache structure.

IntegrationYou access the OLAP Cache Monitor (transaction RSRCACHE) from the SAP Easy Accessscreen, from Business Explorer � BEx Monitor � OLAP: Cache Monitor. From the QueryMonitor screen, you can get to the OLAP Cache Monitor via Cache Monitor.

Settings for the global cache parameters are made in Customizing (see Global CacheParameters [Seite 32]).

The cache-relevant settings that were set for an InfoProvider or query can “overwrite” the globalcache parameters.

� You can set the standard cache mode for an InfoProvider in Customizing. If you changethese settings later the changes do not affect existing queries for this InfoProvider but onlytake effect for future queries created for this InfoProvider.

� You can change the cache mode for a query in the Query Monitor screen (see QueryProperties [Seite 13]).

You can access an overview of the Cache-relevant Data for a query (for example,important time stamps) from the Query Monitor screen via Technical Info (seeTechnical Information [Seite 24]).

FeaturesBy clicking the pushbutton on the left-hand side of the screen, you are able to select whichinformation is displayed in the right-hand screen area. You have the following options:

Cache ParametersDisplay global cache parameters. The display is based on the settings made in Customizing.Example:

Cache parameters

Cache active

Local cache size 100 MB

Global cache size 200 MB

Cache persistence mode Cluster table

Choose Cache Parameter to change any settings. You can find additional information underGlobal Cache Parameters [Seite 32].

You can find additional information on the system valuations specified in the firstcolumn under Technical Information [Seite 24].


BEx Monitor 350 29

Cache Mode FunctionsThe following functions are arranged in pairs in relation to the different cache modes (see CacheMode [Seite 17]).

Cache mode Main Memory Cache with/ Without Swapping

Pushbutton Display

Main Memory Cache-relevant register code for the query run-time objects displayed, buffer status displayed(Shared Memory)

Buffer Objects Buffer objects displayed

You will find additional information under Cache Data Removal and Swapping [Seite35].

Cache mode Cluster / Flat File Cache per Application Server

Pushbutton Display

AppServer Cache-relevant register code for the queryruntime objects displayed


Cache mode Cross-application Server Cluster / Flat File Cache

Pushbutton Display

Cross-AppServer Cache-relevant register code for the queryruntime objects displayed


When you choose Buffer Objects the system displays the buffer content in a hierarchicallystructured query directory.

You can choose from the following display modes:

List Display

Hierarchical Display (standard setting)

You can find additional information under Cache Structure [Seite 38].

Toolbar FunctionsThe following functions are available in the screen toolbar:

General functions


(Refresh)


BEx Monitor 350 30

(Delete) Deletes the cached data in accordance with the cache mode selected.That means that those buffer objects that you chose to display aredeleted. If no selection has been made the system does not delete anydata. This function is useful when testing

Cache Parameter See Global Cache Parameters [Seite 32]

Functions for the cache mode Main Memory Cache with/ Without Swapping


Buffer Monitor Choose Buffer Monitor or Buffer Overview to check, for example,whether enough cache memory is available.

You can use Buffer Monitor to arrive at the Buffer Monitoring Toolscreen, which gives you can overview of the buffer for your applicationserver. In the Buffer Overview screen area, the shared memory rows(buffer: Exp/ImpMem) contain data relevant to the OLAP cache:

� Size: Maximum size of the cache memory in bytes

� Free bytes: Free cache memory in bytes

� Directory: Maximum number of directories, that is, individualpackages in the cache memory

� Dir. Free: Number of free directories

� Efficiency: Degree to which requests can be met, in %, -corresponds to the Hit Ratio in the Buffer Overview

Buffer Overview The Tune Summary screen appears (transaction ST02). The sharedmemory row (buffer: Exp/Imp SHM) contains data of interest for theOLAP cache:

� Hit Ratio: [%] - corresponds to Efficiency in the Buffer Monitor

� Allocated: [kB]

� Free space: [kB] and [%]

� Dir. size: Entries

� Free directory: Entries and [%]

Note that both the number of bytes and the number ofdirectories have a limiting effect. You can find out which ofthese sizes may be critical from the relationship betweenthe respective maximum value and free capacity value.

You will find additional information under The BufferMonitor [Extern]

Functions for the cache mode Main Memory Cache with Swapping (into a file), and thepersistence mode with Flat File Cache



BEx Monitor 350 31

Logical File Name See Logical File Names [Seite 41] (transaction FILE).

Directory Overview The SAP Directories screen appears.

DIR_GLOBAL /usr/<SYSID>/global

Double-click on the physical path to get to the file display. If you havekept to the naming conventions, you can find your physical file underCACHE_<HOST>_<SYSID>_<PARAM_1> orCACHE_<SYSID>_<PARAM_1>.

You can find additional information about defining logical and physicalfile paths and –names under Logical File Names [Seite 41].


BEx Monitor 350 32

Global Cache ParametersUseGlobal cache parameters are the global settings for the OLAP cache (they are valid system-wide).

The following table provides an overview of the different OLAP cache types:

OLAP cache type Description

Cross-transaction cache Cross-transaction application buffer, in which the query navigation statesand query results that were calculated using the OLAP processor arestored on the application server instance. With similar query requests,the OLAP processor can access the data stored in the cache.

Query execution is accelerated when the OLAP processor can read datafrom the cache. This is because the cache can be accessed muchquicker compared with InfoProvider accesses

Local cache Local OLAP processor cache, in which the results calculated by theOLAP processor are stored in the roll area for each session

IntegrationThe global cache parameters are determined centrally during implementation of the BW system.

You can find additional information in the Business Information WarehouseImplementation Guide (IMG), under Reporting-Relevant Settings � GeneralReporting Settings � Global Cache Settings.

It may emerge from the evaluation of data displayed in the OLAP Cache Monitor that the globalcache parameters need to be adjusted to fit system demands. In this case, you are able tochange these settings later.

You can reach the Change View “OLAP Cache Parameters“ View: Detail screen fromthe Cache Monitor screen, via Cache Parameter. Make the desired changes.

FeaturesCache InactiveSetting this parameter means that the cross-transactional cache is deactivated centrally.

Local Size MB: Local Cache Size in MbytesThis parameter determines the memory size of the local OLAP processor cache.

Note that the local cache is always used, including the following cases:

� When the cross-transactional cache was deactivated (see the parameter Cache Inactive).

� When the cache was deactivated for the InfoProvider (for all future queries) or the query(see Cache Mode [Seite 17])


BEx Monitor 350 33

� When it emerges during the runtime that it cannot be cached (see Technical Information[Seite 24] under Cache-Relevant Data: Query Can Use Cache)

Global Size MB: Global Cache Size in MbytesThis parameter determines the maximum memory consumption for all objects in the cross-transaction cache for the cache mode Main Memory Cache With or Without Swapping. You canfind additional information under Cache Mode [Seite 17].

Memory consumption means the amount of memory used by the runtime objects. The actualmemory usage is generally smaller, since the runtime objects are written to the cross-transactional application buffer in compressed form.

The actual size of the cross-transactional cache depends on the size settings of the cross-transactional application buffer: The maximum cache size is the minimum from the Size GlobalMB parameter and the memory size that is actually available in the cross-transactional applicationbuffer.

Persistence ModeThe persistence mode allows you to determine whether and in which form cache data is to bestored:

Cache mode Main Memory Cache with or Without Swapping

The Persistence Mode parameter determines what happens to the data if the maximum size forthe memory is reached. A proportion of the data must either be removed or swapped. A processusing the Least Recently Used (LRU) algorithm determines which data is affected by this.

Cache mode Cluster/ Flat File Cache per Application Server or Cross-application Server

The Persistence Mode parameter determines whether the data is to be stored in a file (flat file) orin a cluster table.

The following table offers an overview of the uses of persistence modes available:

Persistence modes

Persistencemode

Cache mode Main Memory Cachewith or Without Swapping

Cache mode Cluster/ Flat FileCache per Application Server orCross-application Server

Inactive When the cache memory has beenexhausted, excess data is removed,i.e. deleted, from it.

Flat file When the memory space availablefor caching has been exhausted, thedata is swapped out in a file (seeFlat File Name).

The cached data is stored as a file ina directory on the application server orcross-application server in thenetwork (see Flat File Name andOver-arching File).

Cluster table When the memory space availablefor caching has been exhausted, thedata is stored in a cluster table in thedatabase

The cached data is stored as a clustertable in the database. Cluster tablesdiffer in whether they have theapplication server in the key. Thisdepends on the cache mode


BEx Monitor 350 34

If you select the initial entry, in other words, a blank entry and none of the above, thesystem sets the persistence mode to inactive and displays this on the Cache Monitorscreen, in the Cache Parameters screen area.

Flat File Name: Cache Persistence Logical File NameLogical file name under which flat files are stored (see Logical File Names [Seite 41]).

BW_OLAP_CACHE

The following table offers an overview of these parameters and their usage:

Parameter flat file name

Cache mode Meaning

Main Memory Cache with Swapping The flat file name specifies the logical file pathunder which removed data is stored afterswapping

Cluster / Flat File Cache per Application Server If the cache data is held in a file the flat filename specifies the logical file path under whichall the data is stored

Over-arching File: Cache Persistence Logical File NameLogical file path under which data is stored (see Logical File Names [Seite 41]).

BW_OLAP_CACHE_SPAN

The following table offers an overview of these parameters and their usage:

Over-arching file parameter

Cache mode Meaning

Cross-application Server Cluster / Flat FileCache

The flat file name specifies the logical file pathunder which all data is stored. The path is notdependent on the application server


BEx Monitor 350 35

Cache Data Removal and SwappingYou can display information on current memory consumption by clicking on Main Memory for theCache Monitor.

The following table offers an overview of the individual specifications:

Runtime objects table

Specification Meaning

Maximum cache size 200MB by default

Current cache size Sum of the size of all cache structure elementsin KB (see Cache Structure [Seite 38], Bytescolumn)

Current swap size Size of background store (flat file or clustertable) in KB

Cache reserved Ratio between maximum cache size andcurrent cache size in %

Current entries, total Sum of current cache entries and current swapentries

Current cache entries Number of all cache entries (cache structureelements). See Cache Structure [Seite 38]

Current swap entries Number of all entries in the background store

Shared Memory table

Specification Meaning

Buffer poll time Time when buffer capacity was last read

Buffer reserved Degree to which the cache memory is used in% This value is the same as the minimum freebytes and/or free directory entries If you wantto look at these restricting sizes in greaterdetail, choose Buffer Monitor or BufferOverview (see OLAP Cache Monitor [Seite 28])

Buffer capacity cache Specification, what proportion of the sharedbuffer is occupied by the cache

When the capacity of the cache (the maximum cache size) is exhausted but more data is to bewritten to the cache, the following solutions are available:

� Data is displaced (deleted) from the cache. See Cache Mode [Seite 17] Main MemoryCache Without Swapping (1)

� Data is swapped from the cache and stored in a background memory (swap). See CacheMode [Seite 17] Main Memory Cache With Swapping (2)

The following section outlines the basic principle of both procedures using status diagrams:


BEx Monitor 350 36

Principle: Caching with Displacement (Main Memory Cache without Swapping)When data is written to the cache, the entry has the status NEW. (A new write receives thisstatus).

When data is written from the cache, the entry has the status READ. (A new read and writereceives this status).

The LRU replacement mechanism starts as soon as the cache memory capacity has beenexhausted. This checks the status of entries and removes the entry that was last read the longesttime ago.

� When the LRU algorithm comes across an entry with the status READ, it resets it toNEW.

� When the LRU algorithm comes across an entry with status NEW, it makes sure that thisentry is overwritten with the new data. If you need to access the overwritten data againafterwards, this data must be read from the database again (cache miss).

write:

write: read: write:

read:

lru:

lru:

NEW READ

Principle: Caching with Storage in a Background Store (Swap) (Main Memory Cache withSwapping)When data is written to the cache, the entry has the status WRITE. (A new write receives thisstatus).

When data is written from the cache, the entry has the status READ WRITE. (A new read andwrite receives this status).

The LRU replacement mechanism starts as soon as the cache memory capacity has beenexhausted. This checks the status of entries and removes the entry that was last read the longesttime ago.

� When the LRU algorithm comes across an entry with the status READ WRITE, it resets itto READ DIRTY. (A new read receives this status). A new write returns the status of theentry to READ WRITE. The DIRTY flag serves as a “lookout”, making sure that this entryis persisted.


BEx Monitor 350 37

� When the LRU algorithm comes across an entry with the status READ DIRTY, it resets itto DIRTY. The LRU algorithm does the same with entries having the status WRITE thatare not read. (A new read resets the entry to the status READ DIRTY, a new write to thestatus WRITE).

� When the LRU algorithm comes across an entry with status DIRTY, it makes sure thatthis entry is stored in the background memory and indicated as SWAPPED. If the dataneeds to be subsequently accessed again, the data can be read from the backgroundmemory. At the same time, a new caching entry is created having the status WRITE.

After having run through the entries a maximum of two times, the LRU algorithm, by resetting theflags, has found an entry that can be overwritten.

write:

write: read: write:

read:

lru:

WRITE ReadWrite

lru:

lru:

read:

read:

read: write: write:write:lru:

SWAPPED DIRTY ReadDirty


BEx Monitor 350 38

Cache StructureWhen you call up a buffer object in the OLAP cache monitor [Seite 28] the system displaysinformation on the structure elements of the cross-transaction OLAP cache in the right-handscreen area.

Logically, the cross-transactional cache is structured as a hierarchy. This corresponds to Hierarchical Display. For every query request a directory is displayed in an extensive querydirectory that maps the structure elements of the cache on a total of three levels.

Physically, all the elements are stored flat in the cache. Their sequence reflects the order in whichthey were created. This corresponds to List Display.

The following table offers an overview of the logical structuring of information in the OLAP cachemonitor according to the memory ID in the hierarchical display, and the corresponding structureelement:

Mapping between memory IDs and structure elements of the cache

Memory ID Structure element Meaning

Query name (technicalname)

QUERY Table with queries in the cache (directory)

Hierarchies / variables VAR/SH Complex structure for variables in fixed filters andpresentation hierarchies that have beenrequested for the query

FF Complex structure for the entire selection for thequery

Selection / data

SP Storage table: data package, a sequence ofbytes

If you double-click on Query Name or Hierarchy / Variables the Detailed Displaydialog box appears. You get the following information on a query: query Name,Cache Created on, Cache Created By, Cache Last Accessed, Last Accessed By.

These details belong, like the query name itself, to the key for the cache.

If no data is loaded from the cache this is often due to the variable assignment. Youcan use the Detailed Display for Hierarchies / Variables to check this: If variables are part of a fixed filter and therefore cannot be changed for querynavigation, and are filled differently in each case (by manual input or user-definedvariables), the cache for the query has to regenerated every time.

We therefore recommend that you make variables for query navigation changeable.In this way you can achieve the highest possible level of reusability for the cache.

The following graphic shows the relationship between the different OLAP cache structureelements.


BEx Monitor 350 39

Query n

...

Query 2

Query 1

Variables / Hierarchies n...Variables / Hierarchies 2Variables / Hierarchies 1

Global Data

Selection 2...

Selection 1

...Currency 1Currency 2Currency 1

Data Package

QUERY

VAR/SH

FF

SP

OLAP Cache Monitor Display

Column Meaning

Columns in the List Display and in the Hierarchical Display:

Memory ID Cache structure element ID

Swapped Indicator is set when data from the cache isstored in the background store

Read flag Indicator is set when data is read from thecache

Write flag Indicator is set when data was written to thecache and the cache mode is activelypersistent

Dirty flag Indicator is set when data had to be written tothe background store before data removal(cache mode = persistently active)

Directory Query directory

Bytes Cache structure element size in bytes

Buffer ID Cache structure element number sequence inthe memory (also part of flat file name,<PARAM_1>, see Logical File Names [Seite41]

Logical ID Internal logical ID of the Cache structureelement

Columns that are only shown in the List Display:

Parent memory ID Cache structure element ID that, logically, lies


BEx Monitor 350 40

on the next highest hierarchy level

Parent logical ID Internal logical ID of parent element

You can find additional information about the meaning of the individual flags under Cache DataRemoval and Swapping [Seite 35].


BEx Monitor 350 41

Logical File NameUseIn order to save data in a file in the cache modes Main Memory Cache with Swapping, Flat FileCache per Application Server and Cross-application Server (see Cache Mode [Seite 17]), thefollowing specifications must first be set:

� Logical file path defined and assigned a platform-specific physical path.

� Logical file name defined

From the Cache Monitor screen, via Logical File Names, you can access the complex tablemaintenance for logical file paths and -names. The table maintenance consists of several sub-dialogs (views). SAP has already delivered a proportion of the required settings.

You need to check these settings and make any adjustments necessary for your file systembefore you can use the persistent OLAP cache with a flat file as a background store.

Note that these assignments are subject to settings made for the transportconnection. (You need a request that can be transported).

FeaturesThe following section gives an overview of the sub-dialog in the table maintenance using anexample configuration. You can find additional detailed information about the procedure underDefining Logical Path- and File Names [Extern].

You can choose the following sub-dialogs via the Dialog Structure:

Logical File Path DefinitionThe (platform-independent) logical file path is evaluated at runtime to get the (platform-dependent) physical path (see Assignment of Physical Paths to Logical Path).

Two examples of this are:

� Logical file path: BW_OLAP_CACHE

Name: BW OLAP Cache

� Logical file path: BW_OLAP_CACHE_SPAN

Name: BW OLAP cache cross-application server

Assignment of Physical Paths to Logical PathAt runtime, the physical path assigned for the corresponding operating system platform isgenerated from the logical file path.

Since the physical path is platform-dependent, the customer must set it her/himself. The closerthe file is to the application server or, with the cross-application server cache mode, to theapplication server, the faster it can be found and read.

Log(ical) path: BW_OLAP_CACHE


BEx Monitor 350 42

Name: BW OLAP Cache

Syntax group: UNIX (all Unix platforms)

Phys(ical) path: /usr/<SYSID>/global/<FILENAME> (You can choose the path up to<FILENAME> according to your own configuration.)

If you cannot find the syntax group for file names and paths for your operatingsystem, you can define your own (see Syntax Group Definition and Assignment ofOperating System to Syntax Group).

The parameter <FILENAME> must always be set for the physical file name (see Logical FileName Definition, Cross-Client). At runtime, the complete platform-specific file name isautomatically constructed from the physical path and the physical file name.

Logical File Name Definition, Cross-ClientAt runtime, a complete platform-specific file name including the path is generated for a (platform-independent) logical file name.

In the following example, cache data is stored on an application server:

Logical file: BW_OLAP_CACHE (We recommend you use the delivered attributesfor the logical file).

Name: BW OLAP cache files

Phys(ical) file: CACHE_<HOST>_<SYSID>_<PARAM_1>

Data format: BIN (binary)

Application area: BW (Business Information Warehouse)

Log(ical) path: BW_OLAP_CACHE (see Logical File Path Definition). Input help isavailable.

In the following example, cache data is stored cross-application server:Log(ical) file: BW_OLAP_CACHE_SPAN

Name: BW OLAP cache cross-application server

Phys(ical) file: CACHE_<SYSID>_<PARAM_1>

Data format: BIN

Application area: BW

Log(ical) path: BW_OLAP_CACHE_SPAN

Make sure that a particular schema is stipulated for the physical file name:

� For application server storage (see Global Cache Parameters [Seite 32] and Cache Mode[Seite 17] Main Memory Cache with Swapping and Cluster / Flat File Cache per ApplicationServer):

CACHE_<HOST>_<SYSID>_<PARAM_1>

� For cross-application server storage (see Global Cache Parameters [Seite 32] and CacheMode [Seite 17] Cluster / Flat File Cache Cross-application Server):


BEx Monitor 350 43

CACHE_<SYSID>_<PARAM_1>

The following table offers an overview of the variables that are only replaced with actual valuesduring run time:

Variables in physical file names

Variable Meaning

<HOST> Computer-/server name (for SY-HOST).

This variable is not required if the data is stored cross-application server

<SYSID> System ID.

This variable serves to differentiate systems if, for example, two different BWsystems are running on one application server, so that they do not over-writeeach other

<PARAM_1> External parameter 1 (sequence number in HEX)

Defining VariablesIf you want to use variables in physical file names and –paths, you can define these via NewEntries.

Syntax Group DefinitionIf required, you can define a new syntax group for operating systems with the same syntax for filenames and paths via New Entries.

Syntax grp: UNIX

Name: All Unix platforms

Length (of file name)

Extension: Choose this option if the data format for the logical data name is allowedto be the file name extension for this syntax group

Active: Choose this option if the operative system for this syntax group is used inthe available system configuration.

Assignment of Operating System to Syntax GroupIf required, you can assign your operating system to a syntax group via New Entries.

OP system: SunOS

Name: SUN Unix

bex monitor

Documents