how to size sap bw on hana...2. extract script suitable for your db and operating system 3. extract...

SAP & SAP Customer Internal

How To Size SAP BW on HANA

Thomas Becker – TIP In-Memory Platform BW P&QE October 2012

© 2012 SAP AG. All rights reserved. 2

DISCLAIMER

Sizing recommendations apply for certified hardware only. Please contact hardware vendor for suitable hardware configuration.

Note that HANA is constantly being optimized. This might have impact on sizing recommendations, which will be reflected in this document. Therefore, check for the latest version of relevant sizing notes.

Note that the sizing guideline in this document refers to SAP BW on HANA (aka Orange) only. Additional applications running on top of HANA are not covered in this document. Please refer to application specific sizing guidelines for details in that case.


SAP BW on HANA Sizing Elements

SAP BW on HANA sizing consists of

Memory sizing for column store data

Memory sizing for row store data

Memory sizing for objects created during runtime(data load and query execution)

Disk Sizing


Depending on project status, there are three sizing approaches available for HANA:

• QuickSizer Approach• Suitable for customer that start „from scratch“• Some details on data models known (such as largest cubes, DSOs, user numbers, etc.)• Not suitable for customers who plan to migrate an existing BW

• DB specific scripts• Analyze data volumes of an existing BW system, based on database dictionary and statistics

information• Scripts available for Oracle, DBx, MSSQL, MaxDB• OS access / DB admin required to run scripts• Fast (runtime: < 5 mins)

• New: ABAP based sizing report• Analyzes data volume of an existing BW system, based on data samples• DB independent report coming with Support Tools Plug-in• Much more accurate than script based sizing

• Strongly recommended, preferred sizing tool

Sizing Approaches


QuickSizer For SAP HANA

1. Start QuickSizer tool: https://websmp103.sap-ag.de/quicksizer– QuickSizer general help

2. Create new project3. Choose ‘SAP NetWeaver BW powered by SAP HANA’ questionnaire


QuickSizer For SAP HANA (cont.)

4. Read help ‘How to fill in questionnaire’

5. Fill out the relevant sectionso Table 5: Definition of InfoCubes for HANA (relevant for HANA sizing)o Table 6: Definition of DataStore Objects on HANA (relevant for HANA sizing)

All other sections are used for BW Application Server Sizing!


DB specific scripts and sizing information provided by note 1637145

• Unix/Linux and Windows scripts draw information from database dictionary tables to determine size of tables in database (excluding indexes, temp table spaces, etc.)

• Distinction between row store tables and column store tables: different compression factors Scripts deliver two values: row store tables footprint and col store tables footprint. These values need to be fed into sizing formula (see next slide)

• Scripts for DB2 family (DB2, DB4, DB6) take into account source database compression, other scripts don‘t (compression factor has to be applied manually)

• To execute scripts, operating system access with DB administrator privileges is required

• Scripts rely on up-to-date database statistics

• Note: after a system cleanup (deletion of unnecessary data) the database system might not reflect the freed space before table reorganization has been performed

Important: ABAP application server sizing to be done separately!

Sizing with DB Specific Scripts


BW on HANA Sizing Formula

RAM:

Disk:

RAM = ( colstore tables footprint * 2 / 4 + rowstore tables footprint / 1.5 ) * c1) + 50 GB

DISKlog = 1 * RAM

DISKpersistence = 4 * RAM

1) c = source database specific compression factor (only required if sourcedatabase compression is used and source database is other than anyof the DB2 family)


Explanation of constants:

• Uplift for runtime objects

• Average compression factor for col store

• Average compression factor for row store

• Source DB compression factor

• Uplift for HANA caches, additional services, etc.

BW on HANA Sizing Formula: Remarks

RAM = ( colstore tables footprint * 2 / 4 +rowstore tables footprint / 1.5 ) * c1) + 50 GB

RAM = ( colstore tables footprint * 2 / 4 +rowstore tables footprint / 1.5 ) * c1) + 50 GB


BW on HANA Sizing Formula: Procedure and Example

1. Download archive „get_size.zip“ from SAP note 1637145 attachments2. Extract script suitable for your DB and Operating system3. Extract file „load_RowStore_List.sql“ into same folder4. Execute script, result will be stored in file „Extent.txt“5. Read sizing figures from script output:

6. Feed figures into Sizing formula:

Assumption: source DB compressed by factor 1.8

RAM = ( colstore tables footprint * 2 / 4 + rowstore tables footprint) / 1.5 ) * DB compression + 50 GB

= (672 GB * 2 / 4 + 50 GB / 1.5) * 1.8 + 50 GB = 715 GBDiskpersistence = 715 GB * 4 = 2,860 GBDisklog = 715 GB


ABAP report developed to overcome deficiencies of DB scripts:• DB independence• Easier deployment• Distinct size information for different types of tables (InfoCubes, DSOs, PSAs, …)• Extrapolation of memory consumption based on future growth• More sophisticated sizing approaches (e.g. non-active data) coming soon!

Availability• Prerequisite: Support Tools Plug-in ST-PI 2008_1_7xx SP7• Preliminary version based on ST-PI 2008_1_7xx SP6 available with note 1736976• Report: /SDF/HANA_BW_SIZING for full system sizing• Function Module: /SDF/HANA_TABLE_SIZE for single tables or list of tables

Documentation

• Please read documentation attached to note 1736976 for details and tips how to run the report

ABAP Based Sizing


Sizing based on sample data• For each DB table read small set of sample records into ABAP source DB

compression can be neglected!• Determine average record length• Calculate table size based on row count from DB statistics (need to be up to date!)• Automatic distinction between row store and column store tables

Versatile parameterization• Parallel execution: List of tables distributed across multiple processes to speed up

runtime• Different levels of precision (resulting in different sample sizes) – ‘Low’ good enough in

most cases!• Future growth projection

Distinct results• Total sizes for InfoCubes, DSOs, Change Logs and PSA tables are determined• Sizes of individual tables help identify potential for clean-up operations• Automatically applies sizing formula: result contains data + runtime memory

ABAP Based Sizing: How it works …


/SDF/HANA_BW_SIZING: Selection Screen

Number of parallel work processes used forsampling. Note: free dialog porcesses

required for each process!

In detailed list of tables, suppress those withsizes lower than 1 MB.

Precision setting: decreases / increasesdata sample rate. Note: lower precisionmeans less run time!

Enter number of years and amount of growth(absolute in GB or relative in %)

File name for output (relative to SAP workdirectoy, or absolute path name)

Store output to file directly. If not set, a resultscreen will be displayed (wich still can besaved as a file at a later point of time).

Future growth: calculate future HANA requirements for next few years, basedon relative or absolute yearly growth


/SDF/HANA_BW_SIZING: Output Screen / Report (1)

Total source size and number of tablesrelevant for row store

Total size and number of tables relevant forcolumn store

Total number of tables processed

Size and number of tables corresponding toInfoCubes, DSOs, Change Logs, PSAs andMaster Data, number of aggregate tables

Minimum amount of physical memoryfor HANA server (including runtime

memory, memory for additional services, etc.), and disk space.Share of resources for row store and

column store (data ony and data includingdynamic runtime memory)



Sizing recommendations considering futuregrowth with detailed sizing requirements for

each year (only if future growth was selected)

System information, incuding report run time, parameter settings, operating and DB system

information and SAP BW release



Detailed size information for each row storetable (sorted by size descending). Large tables may be good candidates for data

cleansing to reduce memory requirements.

Report shows ABAP size and size in HANA memory

Detailed size information for each columnstore table (sorted by size descending).

Helps to identify good candidates for datacleansing to reduce memory requirements

(e.g. very large PSA tables, etc.)

Distinuishes between PSA (P) andChange Log (C) tables


Outlook: Not-Active Data in HANA

hot

warm

cool

Data is read/written frequentlyIn Memory, additional memory required for dynamic objects (merge, intermediate results, etc.)

Mostly read access – lookups, transformations, etc.In Memory, no additional memory required for dynamic objects

Infrequent access – no need to keep in memory all the timeOn disk, loaded to memory only on demand, good candidate for displacement if memory runs short

Non-active data concept available with NetWeaver BW SPS8 and HANA SP5

Will be reflected in next version (V1.3) of Sizing Report – coming soon!• Optimized usage of HANA memory resources• Modeling of LSA concepts (Corporate Memory, …)

NOTE: Due to the complexity of the sizing algorithm, only the ABAP report will be able to consider non-active data. Database scripts and QuickSizer will NOT be changed to reflect this feature!


If a single HANA node cannot accommodate data due to limited memory, data has to be distributed across multiple nodes (scale-out).

Symmetric solution: 1 master node, n slave nodes, all on identical hardware

Master node will handle system load and transactional load: ABAP system tables and general operational data of the BW are stored on the master node. DDL statements are executed on this node, global locks are acquired here.

Slave nodes will handle OLAP queries as well as loading/staging/activation/merging.BW data (master data + cubes/DSOs/PSAs) is distributed across all slaves. This ensures a balanced utilization of the available CPU and memory resources.

Easy extensibility: When a new slave is added, BW data can quickly be reorganized to fit the new system. More information here: http://help.sap.com/hana/hana_db_part_en.pdf

Optional: stand-by node(s) can take over in case of node failure

BW on HANA Sizing: Scale Out


HANA BW Scale Out Concept:Overview 1/2 – Setup with 512 GB Nodes

Master Node512 GB of memoryholds system tables, all row tablesDDL execution, global locks, database metadata

Up to 14 Slave Nodes512 GB of memory eachmaster data tables will be distributed across slave nodesfact, DSO and PSA tables are partitioned and distributed

Standby Node(s) (optional)512 GB of memorycan take over role of master node or of one of the slaves for failover

System

Tables

Master

BW Data

Slave 1

BW Data

Slave 2

BW Data

Slave n

…

Standby


HANA BW Scale Out Concept:Overview 2/2 – Setup with 1 TB Nodes*

Master Node1 TB of memoryholds system tables, all row tablesDDL execution, global locks, database metadata

Up to 14 Slave Nodes1 TB of memory eachmaster data tables will be distributed across slave nodesfact, DSO and PSA tables are partitioned and distributed

Standby Node(s) (optional)1 TB of memorycan take over role of master node or of one of the slaves for failover

System

Tables

Master

BW Data

Slave 1

BW Data

Slave 2

BW Data

Slave n

…

Standby

* For Hardware vendors that support it


HANA BW Scale Out Concept:Pure High Availability Solution with a Standby Node

Worker Node512 GB or 1 TB of memorysingle worker node

Standby Nodesame amount of memory as worker nodecan take over role of worker node for failover

Worker

Standby


HANA BW Scale Out Concept:Exception for systems with a single slave node

Master Nodeholds system tables, all row tablesDDL execution, global locks, database metadata

Slave Nodeholds master data, fact, DSO and PSA tables

Standby Node (optional)can take over role of master or slave node for failover

System Tables

Master

BW Data

Slave

Standby

Please see note 1702409 why this setup is not recommended.

For special requirements, master data tables may optionally be moved to the master node. This can lead to performance degradation. Master data tables may be moved manually (there is no official tool support) or with the help of a consulting note.

Thank You!Contact information:

Dr. Thomas BeckerQuality ManagerTIP In-Memory PlatformBW Performance & Quality [email protected]


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, Excel, Outlook, and PowerPoint are registered trademarks of Microsoft Corporation. IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x, System z, System z10, System z9, z10, z9, iSeries, pSeries, xSeries, zSeries, eServer, z/VM, z/OS, i5/OS, S/390, OS/390, OS/400, AS/400, S/390 Parallel Enterprise Server, PowerVM, Power Architecture, POWER6+, POWER6, POWER5+, POWER5, POWER, OpenPower, PowerPC, BatchPipes, BladeCenter, System Storage, GPFS, HACMP, RETAIN, DB2 Connect, RACF, Redbooks, OS/2, Parallel Sysplex, MVS/ESA, AIX, Intelligent Miner, WebSphere, Netfinity, Tivoli and Informix are trademarks or registered trademarks of IBM Corporation.Linux is the registered trademark of Linus Torvalds in the U.S. and other countries.Adobe, the Adobe logo, Acrobat, PostScript, and Reader are either trademarks or registered trademarks of Adobe Systems Incorporated in the United States and/or other countries.Oracle and Java are registered trademarks of Oracle and/or its affiliates.UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group.Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame, and MultiWin are trademarks or registered trademarks of Citrix Systems, Inc.HTML, XML, XHTML and W3C are trademarks or registered trademarks of W3C®, World Wide Web Consortium, Massachusetts Institute of Technology.

© 2012 SAP AG. All rights reserved.

SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, SAP BusinessObjects Explorer, StreamWork, and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries.

Business Objects and the Business Objects logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius, and other Business Objects products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of Business Objects Software Ltd. Business Objects is an SAP company.

Sybase and Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere, and other Sybase products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of Sybase, Inc. Sybase is an SAP company.

All other product and service names mentioned are the trademarks of their respective companies. Data contained in this document serves informational purposes only. National product specifications may vary.

The information in this document is proprietary to SAP. No part of this document may be reproduced, copied, or transmitted in any form or for any purpose without the express prior written permission of SAP AG.


© 2012 SAP AG. Alle Rechte vorbehalten.

Weitergabe und Vervielfältigung dieser Publikation oder von Teilen daraus sind, zu welchem Zweck und in welcher Form auch immer, ohne die ausdrückliche schriftliche Genehmigung durch SAP AG nicht gestattet. In dieser Publikation enthaltene Informationen können ohne vorherige Ankündigung geändert werden.Die von SAP AG oder deren Vertriebsfirmen angebotenen Softwareprodukte können Softwarekomponenten auch anderer Softwarehersteller enthalten.Microsoft, Windows, Excel, Outlook, und PowerPoint sind eingetragene Marken der Microsoft Corporation. IBM, DB2, DB2 Universal Database, System i, System i5, System p, System p5, System x, System z, System z10, System z9, z10, z9, iSeries, pSeries, xSeries, zSeries, eServer, z/VM, z/OS, i5/OS, S/390, OS/390, OS/400, AS/400, S/390 Parallel Enterprise Server, PowerVM, Power Architecture, POWER6+, POWER6, POWER5+, POWER5, POWER, OpenPower, PowerPC, BatchPipes, BladeCenter, System Storage, GPFS, HACMP, RETAIN, DB2 Connect, RACF, Redbooks, OS/2, Parallel Sysplex, MVS/ESA, AIX, Intelligent Miner, WebSphere, Netfinity, Tivoli und Informix sind Marken oder eingetragene Marken der IBM Corporation.Linux ist eine eingetragene Marke von Linus Torvalds in den USA und anderen Ländern.Adobe, das Adobe-Logo, Acrobat, PostScript und Reader sind Marken oder eingetragene Marken von Adobe Systems Incorporated in den USA und/oder anderen Ländern.Oracle und Java sind eingetragene Marken von Oracle und/oder ihrer Tochtergesellschaften.UNIX, X/Open, OSF/1 und Motif sind eingetragene Marken der Open Group.Citrix, ICA, Program Neighborhood, MetaFrame, WinFrame, VideoFrame und MultiWin sind Marken oder eingetragene Marken von Citrix Systems, Inc.

HTML, XML, XHTML und W3C sind Marken oder eingetragene Marken des W3C®, World Wide Web Consortium, Massachusetts Institute of Technology.SAP, R/3, SAP NetWeaver, Duet, PartnerEdge, ByDesign, SAP BusinessObjects Explorer, StreamWork und weitere im Text erwähnte SAP-Produkte und -Dienstleistungen sowie die entsprechenden Logos sind Marken oder eingetragene Marken der SAP AG in Deutschland und anderen Ländern.Business Objects und das Business-Objects-Logo, BusinessObjects, Crystal Reports, Crystal Decisions, Web Intelligence, Xcelsius und andere im Text erwähnte Business-Objects-Produkte und -Dienstleistungen sowie die entsprechenden Logos sind Marken oder eingetragene Marken der Business Objects Software Ltd. Business Objects ist ein Unternehmen der SAP AG.

Sybase und Adaptive Server, iAnywhere, Sybase 365, SQL Anywhere und weitere im Text erwähnte Sybase-Produkte und -Dienstleistungen sowie die entsprechenden Logos sind Marken oder eingetragene Marken der Sybase Inc. Sybase ist ein Unternehmen der SAP AG.Alle anderen Namen von Produkten und Dienstleistungen sind Marken der jeweiligen Firmen. Die Angaben im Text sind unverbindlich und dienen lediglich zu Informationszwecken. Produkte können länderspezifische Unterschiede aufweisen.Die in dieser Publikation enthaltene Information ist Eigentum der SAP. Weitergabe und Vervielfältigung dieser Publikation oder von Teilen daraus sind, zu welchem Zweck und in welcher Form auch immer, nur mit ausdrücklicher schriftlicher Genehmigung durch SAP AG gestattet.

how to size sap bw on hana...2. extract script suitable for your db and operating system 3. extract...

Documents