SAP-BW Performance Monitoring With BW Stat

May 28, 2016 | Author: therdsakc | Category: Types, Presentations
Share Embed Donate


Short Description

SAP-BW Performance Monitoring with BW Statistics...

Description

Know How Network: SAP BW Performance Monitoring with BW Statistics Ron Silberstein Platinum Consultant - Business Intelligence Netweaver RIG US SAP Labs, LLC

Agenda

BW Statistics Overview Monitoring with BW Statistics Reports Performance Monitoring Tools Detecting Optimization Potential Setting Up BW Statistics

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 2

2

Agenda

BW Statistics Overview

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 3

3

BW Statistics: Facts

4

SAP-delivered tools for analyzing and tuning system performance Delivered SAP Business Content Usage of tools requires Business Content installation Content Delivery Consists of: MultiCube InfoCubes Queries Workbook (New one as of BW 2.0B) InfoSources Transfer Structures, Transfer Rules, Update Rules InfoPackages, InfoPackage Groups Role

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 4

BW Statistics: Facts (continued) Maintenance requires scheduling data loads and monitoring Delta load functionality delivered Transaction Data Master Data

One of several sources of performance monitoring information. Others include: ST03 > Expert Mode > BW system load Table RSDDSTAT Function Module RSDDCVER_RFC_BW_STATISTICS

Can be utilized when building aggregates System can “propose” aggregates from BW stats data

Object metadata of BW Statistics is transportable

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 5

5

BW Statistics: Data Flow Architecture

6

Turn on/off logging for each InfoCube!

Table RSDDSTAT

Queries Queries

InfoInfoSource Source

Update Update rules rules

aggregates aggregates

InfoInfoSource Source

Update Update rules rules

InfoInfoSource Source

Update Update rules rules

Table RSDDSTATAGGR

warehouse warehouse management management

Table RSDDSTATWHM

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 6

BW Statistics MultiProvider Analysis by queries (in workbooks)

Enable Capture of Statistical Performance Data Choose Tools > BW Statistics for InfoProviders

Can be turned on/off for OLAP/WHM

You can store new settings permanently

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 7

You can delete old data from database tables

7

Consequences

8

If the BW statistics are turned on, the systems collects data. These data is inserted in different tables: Tables, which are filled by ‚OLAP‘:

Tables, which are filled by ‚WHM‘:

RSDDSTAT

RSDDSTATWHM

RSDDSTATAGGR

RSDDSTATCOND

RSDDSTATAGGRDEF

RSDDSTATDELE, ...

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 8

Agenda

Monitoring with BW Statistics Reports

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 9

9

Key Delivered Queries (BW Statistics Workbook) Use of OLAP for Each InfoCube (Time) Components of runtime per query

Use of OLAP for Each InfoCube (Records) Records selected / transferred

Query Runtime User, query, # of navigations, components of runtime

Use of OLAP by Date Use of OLAP according to time of day Mean Time Per Query Mean time per navigation

Data Load: Utilizing WHM per InfoSource

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 10

10

Example: SAP Delivered Graph (Workbook)

Total runtime (blue) by hour and Total number of records (red) by hour

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 11

11

Customer Report Example: Usage by Month (Cube)

# of Navigations

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 12

12

Customer Report Example: Usage by Month (User)

Enterprise view # of navigations by user

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 13

13

Example: Usage by Month (User) for a Cube

InfoCube view # of navigations by user

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 14

14

Customer Example: Aggregate Use Per Cube

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 15

15

Customer Example: Aggregate Use Per Cube

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 16

16

Agenda

Performance Monitoring Tools

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 17

17

Overview: Reporting Performance Analysis Tools 18

Table RSDDSTAT

BW Statistics

BW Workload Analysis – ST03N

Collecting information from table RSDDSTAT Function module RSDDCVER_RFC_BW_STATISTICS 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 18

ST03 – Workload Analysis – BW Specific

Expert mode Expert mode Administrator Service engineer

BW System Load

2

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 19

1

19

ST03 - WorkLoad By InfoCube Highest contributor in Total Run Time per InfoCube

20 High % DB Time

Expert Mode! High ratio: Selected / transferred records

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 20

Using Table RSDDSTAT

21

In transaction SE16, display table RSDDSTAT

Number of entries

INFOCUBE

to

QUERYID

to

QRUNTIMECATEGORY

to

QNAVSTEP

to

QDBSEL

to

QDBTRANS

to

QTIMEDB

to

STARTTIME

to

Width of output list Maximum no. of hits

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 21

YYYYMMDDHHmmss

Settings of Table RSDDSTAT Choose Settings

22

User parameters (F6) User-Specific Settings Repository Infosystem

INFOCUBE QUERYID QRUNTIMECATEGORY QNAVSTEP QDBSEL QDBTRANS QTIMEDB STARTTIME

Data Browser

ITS

Chan...

Output list Width of output list Maximum no. of hits Display max.n.of hits possible ALV-Grid ALV-List SE16 standard list Keyword Field name Field text

Width of output list Maximum no. of hits

Check conversion exits

Set flag to display query names 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 22

Important Fields in Table RSDDSTAT (1) Query name

Runtime Category

Database time

23

Records selected on the database

Records transferred from the database to the OLAP

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 23

Agenda

Detecting Optimization Potential

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 24

24

Analyzing Reporting Performance Issues Building Aggregates: Primary technique for tuning reporting performance Analyze to determine if building aggregates will help

Different Starting Points: Analyze specific InfoProviders OR Find problem queries, then analyze them

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 25

25

Analyzing Reporting Performance Issues

Approach: Find InfoCubes of concern & consider aggregates

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 26

26

BW Statistics or ST03N: Find InfoCubes for Analysis 27

Sort by mean overall time to find InfoCubes with queries having the highest runtimes

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 27

Stastistics for Queries BEx Analyzer

3

Front-end

Query execution

OLAP processor

Application Server

2

Database Server

InfoCube Data

Amount of data transferred to the front-end?

Number of rows transferred to the application server?

1

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 28

28

Web Browser

Number of rows selected on the database?

Aggregates

Front-end time?

OLAP time?

Database time?

Indicators That Aggregates Will Help Look for many more records selected than necessary!

Infocube

29

One Key Indicator: Ratio records selected / records transferred > 10

Number of records read on the database (QDBSEL)

AND

Records transferred after being summarized on the database (QDBTRANS)

Other Key Indicator: database time for queries > 30% of total query runtime

Total query runtime (QRUNTIMECATEGORY) Start

2s

4s

6s

8s

10s

12s

14s

Database time (QDBTIME) 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 29

16s

18s

20s End

Analysis of BW Statistics at InfoCube Level (1) 30

Database time ~40% of total runtime (6771 of 17 025 seconds)

Aggregates will probably improve query performance

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 30

Ratio records selected / records transferred: 24

Analysis of BW Statistics at InfoCube Level (2) 31

Database time ~3,5% of total runtime (1011 of 28 540 seconds)

Aggregates will not improve query performance

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 31

Ratio records selected / records transferred: 32

Analysis of BW Statistics at InfoCube Level (3) 32

Database time ~90% of total runtime (1420 of 1572 seconds)

Aggregates will not improve query performance

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 32

Ratio records selected / records transferred: 1.1

Analyzing Reporting Performance Issues 33

Approach: Find queries of concern, then analyze them

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 33

Indicators That Aggregates Will Help Look for many more records selected than necessary!

Infocube

29

One Key Indicator: Ratio records selected / records transferred > 10

Number of records read on the database (QDBSEL)

AND

Records transferred after being summarized on the database (QDBTRANS)

Other Key Indicator: database time for queries > 30% of total query runtime

Total query runtime (QRUNTIMECATEGORY) Start

2s

4s

6s

8s

10s

12s

14s

Database time (QDBTIME) 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 34

16s

18s

20s End

Using BW Statistics: Find Queries for Analysis 35

Utilizing OLAP per Query

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 35

Sort by overall time to find queries that have the highest total runtime

Analysis Techniques: Key Components of Runtime 1 36

Utilizing OLAP per Query

Database time ~60% of total runtime (2.798 of 4.685 seconds)

Aggregates will improve query performance

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 36

Ratio records selected / records transferred: 54

Analysis Techniques: Key Components of Runtime 2 37

Database time ~89% of total runtime (1064 of 1162 seconds)

Ratio records selected / records transferred: 2.1

Aggregates will not improve query performance 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 37

Performance Concerns: if not aggregates, then what? 38

Resource Constraints? CPU speed Number of CPUs Available memory I/O throughput, number of controllers

Database Statistics – Are they complete and up-to-date? Use of OLAP Cache Query Design – Start at summarized level, drill down for details? Design Considerations InfoCube Partitioning MultiProvider Line Item Dimension ODS – InfoCube – Report to Report Interface

Web Applications Compression Precalculation 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 38

Agenda 39

Setting Up BW Statistics

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 39

Enable Capture of Statistical Performance Data

40

Choose Tools > BW Statistics for InfoProviders

Can be turned on/off for OLAP/WHM

You can store new settings permanently

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 40

You can delete old data from database tables

Customizing: Transfer Business Content DataSources 41

Where? Transaction SBIW

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 41

Customizing: Transfer Business Content DataSources 42

All Components Under “TCT” Must Be Selected for Transfer

Where? Transaction SBIW, Transfer Business Content DataSources 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 42

Customizing: Transfer Business Content DataSources 43

Choose “Select Delta”, Then “Transfer DataSources 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 43

Replicate DataSources (Self Source System) 44

Where? Administrator Workbench > Modeling > Source Systems > Select Self Source System > Context Menu 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 44

Business Content Installation: Group Objects 45

Where? Administrator Workbench > Business Content > Grouping Options

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 45

Business Content Installation: Select InfoProviders 46

Drag and Drop BW Statistics InfoProviders into Collected Objects Workspace

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 46

Business Content Installation: Check & Install

Ensure That All Objects in Tree Are Selected For Installation (or are already active), then Choose “Install” 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 47

47

Business Content Installation: Collect Object Superset 48

Repeat Object Grouping Steps, But Now Choose Grouping Option All Objects In Dataflow Before and After 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 48

Business Content Installation: Check Objects & Install 49

Spot Check Objects: Expand Update Rules, then Continue Expansion to InfoPackages Install

Ensure All Objects Are Selected for Installation (or active) 2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 49

Post-Content Installation: Check InfoPackages 50

Where? Admin WB > Modeling > InfoSources: Check for Existence of InfoPackages

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 50

Post-Content Installation: Check Data Flow Diagram 51

Where? Admin WB > Modeling > InfoProvider > Technical Content > BW Statistics > MultiProvider for BW Statistics > Context Menu > Show Data Flow

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 51

Post-Content Installation: Check Data Flow Diagram 52

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 52

BW Statistics Setup: Data Loads 53

Check delivered InfoPackage groups; Utilize and modify as needed. Consider building process chains to automate data loads. InfoPackage Scheduler: perform data loads for all master data InfoSources under the Technical Content InfoArea and the BW Metadata InfoArea InfoPackage Scheduler: Perform delta initializations for all transaction data InfoPackages under the Technical Content InfoArea InfoPackage Scheduler: Schedule regular delta data loads for all transaction data InfoPackages under the Technical Content InfoArea InfoPackage Scheduler: Schedule regular data loads for all master data InfoPackages under the Technical Content InfoArea and the BW Metadata InfoArea Regularly monitor data loads and ensure timely data availability.

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 53

BW Statistics Setup: Assign Technical Content Role 54

The Technical Content Role (SAP_BWC_TCONT ) is activated via the BW statistics business content installation. The delivered BW Statistics Workbook is (by default) contained within the Technical Content Role (SAP_BWC_TCONT ). Use transaction PFCG to assign the Technical Content Role (SAP_BWC_TCONT) to any users who should have easy access to the BW Statistics Workbook. Important OSS notes (BW statistics) : 422264, 309955, 551552

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 54

Important Resource 55

How-to-guide: “How to Setup BW Statistics”: Valid for versions 2.0B, 2.1C, 3.0B and 3.1Content

Available at http”//service.sap.com/BW > Services & Implementation > How to Guides > Guide list BW 3.0

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 55

Questions? 56

Q&A

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 56

Copyright 2003 SAP AG. All rights reserved No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice. Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors. Microsoft®, WINDOWS®, NT®, EXCEL®, Word®, PowerPoint® and SQL Server® are registered trademarks of Microsoft Corporation. IBM®, DB2®, OS/2®, DB2/6000®, Parallel Sysplex®, MVS/ESA®, RS/6000®, AIX®, S/390®, AS/400®, OS/390®, and OS/400® are registered trademarks of IBM Corporation. ORACLE® is a registered trademark of ORACLE Corporation. INFORMIX®-OnLine for SAP and Informix® Dynamic ServerTM are registered trademarks of Informix Software Incorporated. 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 of Citrix Systems, Inc. HTML, DHTML, XML, XHTML are trademarks or registered trademarks of W3C®, World Wide Web 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 for technology invented and implemented by Netscape. SAP, SAP Logo, R/2, RIVA, R/3, SAP ArchiveLink, SAP Business Workflow, WebFlow, SAP EarlyWatch, BAPI, SAPPHIRE, Management Cockpit, mySAP.com Logo and mySAP.com are trademarks or registered trademarks of SAP AG in Germany and in several other countries all over the world. All other products mentioned are trademarks or registered trademarks of their respective companies.

2003 SAP Labs, LLC, SAP Professional Journal Conference, Ron Silberstein 57

View more...

Comments

Copyright ©2017 KUPDF Inc.
SUPPORT KUPDF