Professional Documents
Culture Documents
1
SAP Systems at OfficeMax
Hazelton, McCalla,
Decentralized Las Vegas
Pennsylvania Alabama
Warehouse Management
Systems WM3
WM1 WM3
WM2 WM3
•Vendor Shipments
•Deliveries made
•Deliveries due
•Inv Adjustments
•Master Data
Shaker Heights, WA ME
Ohio
MT ND VT
MN NH
OR WI NY MA
Retail
WM3
S NC
TN
AZ NM OK AR SC
MS AL GA
TX LA
FL
Shaker Heights,
Shaker Heights,
•Campaign Analysis Ohio
Ohio
WM3
CRM BW/SEM
WM3
•Financial Simulation
(transactional cube)
•Market Basket
(data mining) Business Intelligence
Customer Call Center •Customer Analysis
(Cleveland, Dallas) and Modeling
2
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
4
Application Areas & InfoProviders
INVENTORY 12 76
PURCHASING 2 4 2
FINANCE 6 4
HR 1 2
CRM 1 3 2
SEM 6
STAFFWORKS 3 3 3
BW STATISTICS 6
Others 5 4
5
Performance Statistics for Loads
Data Process Records Runtime Streams
Master Data
Change run 20000 3 hours 8
Alignment
6
BW Users - Yearly
680
700
600
504
500
400
300 Users
195
200 149
100
0
Year- Year- Year- Year
2004 2003 2002 2001
7
BW Users - Daily
250
226 222
200
178
150 141
150
121 117
106 Users
100 96
91
98 99
92
57
52
50
0
8/16/2004 8/23/2004 8/30/2004 8
2
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
10
Modeling – Performance Aspects - 2
Line Item Dimensions & Cardinality
11
Modeling – Performance Aspects - 3
z Logical Partitioning – Yearly cubes
X Parallel queries run on basic infocubes
X Change run can be run parallel
12
Modeling – Performance aspects - 4
Run Program SAP_INFOCUBE_DESIGNS for existing Infocubes
Compression ratio
13
Modeling – Performance Aspects - 5
z Physical Partitioning
z Aggregates
z Navigational attributes
z Database partitions and
secondary Indexes on ODS.
z Dimensions with more than one Characteristic
14
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
16
Data Load Performance – Number Range Buffering - 1
z When loading large quantities of data into an InfoCube, the number range buffer
should be increased for the dimensions that are likely to have a high number of
values
z Use Function module RSD_CUBE_GET to Identify the object name of the
dimension.
17
Data Load Performance – Number range buffering - 2
z Use Transaction SNRO ->EDIT-> Setup Buffering->Main memory
z Never set buffering for package dimension
18
Data Load Performance – Load Balancing - 1
Use transaction SMLG and create logon group
BW system
19
Data Load Performance – Load balancing - 2
z Change RFC settings in the source system
z Use transaction SM59
Source system
20
Data Load Performance
z Drop Unused Partitions
z Use program SAP_DROP_EMPTY_FPARTITIONS
X Avoid Long execution times for Index rebuild
21
Data Load Performance
z Drop and recreate source twin Structures
X Program LMCSWTOP set flag FLG_DCT = ‘X’
z Drop and Recreate Bitmap indexes on F table
z Use Parallel Processing
z ODS loads & activation parallel
z Check Update / Transfer rules
z Index on source tables for full loads with selection
z Define PSA Part. size using transaction RSCUSTV6
X Minimum no. of records in a PSA partition
X Use program SAP_PSA_PARTITION_COMPRESS for existing PSA
tables
z Compress Cubes and aggregates regularly
22
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
24
Change run Performance - 2
z Function module RSCDS_CONDENSE_CUBE
Aggregate name
25
Change run Performance – Threshold & Block size
Transaction RSCUSTV8
26
Change run Performance - Setting Change run Parallel -1
Create an RFC server group using transaction RZ12
27
Change run Performance - Setting Change run Parallel -2
z After the RFC server group is defined use report SAP_RSADMIN_MAINTAIN to create
the object entries CR_RFCGROUP in the table RSADMIN
z OSS Note – 534630 Parallel Processing of Change run
z Parallel Processing of Change run Increases memory consumption
28
Change run Performance – Monitoring - 1
Use Table RSDDSTATAGGR to find the run times
29
Change run Performance – Monitoring - 2
Program RSDDS_CHANGERUN_MONITOR
30
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
} Degenerated Indexes
32
Query Performance - 2
z Build Aggregates
z OLAP Cache
z Pre Calculated Value Sets
z Pre Calculated Web Templates
z Logical Partitioning - Multi Providers
z Physical Partitioning
z Look for degenerated Indexes
z Compress Infocubes regularly
33
Query Performance – OLAP Cache
OLAP Cache Settings using transaction RSCUSTV14
34
Query Performance – Pre Calculated Web Templates
35
Query Performance – Pre-Calculated Valuesets
36
Query performance Tuning Techniques - 1
Switch on Statistics
37
Query Performance Tuning Techniques - 2
ST03N Expert mode
38
Query Performance Tuning Techniques - 3
z Run Transaction SM66
z Identify the process id
z Run transaction ST04 and find the process in Oracle session
z Get the execution Plan
39
Query Performance Tuning Techniques - 4
40
Query Performance Tuning Techniques - 5
z Read mode (Table RSRREPDIR)
z Tables RSDDSTAT / RSDDSTATAGGR
z Cache Monitor(RSRCACHE)
z Function Module
RSDDCVER_RFC_BW_STATISTICS
z BW Stats Cubes
z RSRT
z RSRTRACE
41
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
z Aggregates Rollup
¾ Compress Cubes & aggregates regularly
¾ Rebuild Indexes on Etable of aggregates
¾ Build statistics on Aggregate tables
¾ Include partitioning Characterstic in the aggregates
¾ Do not build stats on the F table of the cube
z Compression Performance
¾ Make sure P index exists on the Etable of Cube & aggregates
¾ Aggregates Ftable size
43
Aggregate Rollup & Compression Performance
z Aggregates Rollup
X Compress Cubes & aggregates regularly
X Rebuild Indexes on E table of aggregates
X Build statistics on Cubes & Aggregates
X Include partitioning Characteristic in the aggregates
X Do not build stats on the F table of the cube if the table is huge
z Compression Performance
X Make sure P index exists on the E table of Cube & aggregates
X Aggregates F table size
44
Agenda
BW Technical Overview
Modeling
Extraction
Changerun
Queries
Activate Stats
46
SEM BPS Performance - 2
z Database Parameter Settings ( OSS Note : 358921)
z Executing Specialized functions faster than General
functions (eg:- Copy function instead of writing
formula to copy)
z Program UPC_BW_SELECTION_DEBUG can be used
to do the analysis of read access in SEM-BPS
z Save the results after ending a planning session
before starting another transaction since the memory
is released only after the save or exiting the session.
z Compress the Infocubes
z Build aggregate if possible
z Code Formula functions effectively avoiding multiple
loops 47
Summary
z Check Oracle Parameter & Basis Parameter
Settings
z Deploy Parallel Processing wherever possible
z Look for degenerated Indexes
z Use transactions ST03N / ST04 / ST02
z ROIDOCPRMS table settings for Load Performance
z Make use of Load balancing
z Use OLAP Cache
z Use Pre-Calculated Value sets and Web Templates
48
Questions?
Q&A
49
Feedback
Please complete your session evaluation.
Be courteous — deposit your trash,
and do not take the handouts for the following
session.
Thank You !
50
SAP Developer Network
Look for SAP TechEd ’04 presentations and videos on
the SAP Developer Network.
Coming in December.
http://www.sdn.sap.com/
51
Copyright 2004 SAP AG. All Rights
Reserved
No part of this publication may be reproduced or transmitted in any form or for any purpose
z
without the express permission of SAP AG. The information contained herein may be changed
without prior notice.
z Some software products marketed by SAP AG and its distributors contain proprietary software
components of other software vendors.
z Microsoft, Windows, Outlook, and PowerPoint are registered trademarks of Microsoft
Corporation.
z 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, and Informix are trademarks or registered trademarks of IBM Corporation in the
United States and/or other countries.
z Oracle is a registered trademark of Oracle Corporation.
z UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.
z Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin are
trademarks or registered trademarks of Citrix Systems, Inc.
z HTML, XML, XHTML and W3C are trademarks or registered trademarks of W3C®, World Wide
Web Consortium, Massachusetts Institute of Technology.
z Java is a registered trademark of Sun Microsystems, Inc.
z JavaScript is a registered trademark of Sun Microsystems, Inc., used under license for
technology invented and implemented by Netscape.
z MaxDB is a trademark of MySQL AB, Sweden.
z SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver and other SAP products and 52
services mentioned herein as well as their respective logos are trademarks or registered
trademarks of SAP AG in Germany and in several other countries all over the world All other