Volume19 En

8/3/2019 Volume19 En

1/82

11gD A T A B A S E

Oracle SAPfor

Oracle database for S

is safe, reliable and

scalable

No.19 Oracle for SAP, May 20

www.oracle.com/sap

T E C H N O L O G Y U P D A T E

N E WRequest Exadata V2-Sun

Oracle Database Machinefor your SAP implementation NOW!


2/82

Oracle Exadata Storage Servers combine Oracles smart storage software and Oracles industry-standard Sun hardware to deliver

the industry's highest database storage performance. To overcome the limitations of conventional storage, Oracle Exadata Storage

Servers use a massively parallel architecture to dramatically increase data bandwidth between the database server and storage.

In addition, smart storage software offloads data-intensive query processing from Oracle Database 11gservers and does the query

processing closer to the data. The result is faster parallel data processing and less data movement through higher bandwidth

connections. This massively parallel architecture also offers linear scalability and mission-critical reliability.

Exadata Smart Flash Cache

Now with the newest release of the Oracle Exadata Storage Server, you can also achieve extreme performance for transaction

processing and consolidated mixed application workloads. Exadata Smart Flash Cache addresses the disk random I/O bottleneck

problem by transparently moving hot data to Oracles Sun FlashFire cards. You get ten times faster I/O response time and use ten

times fewer disks for business applications from Oracle and third-party providers.

Exadata Hybrid Columnar Compression

The most advanced data compression technologies are included in Oracle Exadata Storage Servers. Oracle Exadata Hybrid Columnar

Compression can, on average, reduce the size of Data Warehousing tables by 10x, and archive tables by 50x. This offers significant

savings on disk space for primary, standby and backup databases, and will improve the performance of Data Warehousing queries.

Benefits for SAP Customers

Extreme Performance for Business WarehousesImproves both query performance and concurrency by offloading intensivequery processing and data mining scoring from database servers and bringing it closer to the data.

Extreme Performance for OLTP ApplicationsAddresses the disk random I/O bottleneck problem by transparently moving hotdata to Sun FlashFire. You get ten times faster I/O response time and use ten times fewer disks for business applications from

Oracle as well as third-party providers.

Extreme Performance for Mixed WorkloadsAllows you to consolidate data warehousing, transaction processing, and mixedworkloads on a massively parallel server grid and eliminate the cost of third-party specialty hardware, security, and management

solutions.

UNVEILING

ORACLE EXADATA V2

The Worlds First OLTP Database Machine

With Sun FlashFire Technology


3/82

ORACLE EXADATA V2The Worlds First OLTP Database Machine

With Sun FlashFire Technology

Operating System and version?

Database version?

Database size?

SAP Application Modules and release level?

Number and type of application servers running

on the existing database server?

Number and type of dedicated application servers?

Yes / No we are using BI Accelerator?

Name:

Company:

Address:

E-mail:

Please fill out the Oracle Exadata V2 for SAP request form and send (or fax) it to Faxnr.:

+49 6227 8398 199

To: Dr. Christian Graf (SAP AG) and Mr. Gerhard Kuppler (Oracle)

Regarding: Exadata V2 Sun Oracle Database Machine

Dear Sirs;

The purpose of this request is to express our interest in adopting Exadata V2 Sun Oracle

Database Machine for our SAP Applications and we are asking you for its immediate

release, once the Oracle Database 11g Release 2 for SAP is available.

Send this request to [email protected] and [email protected]

(or fax this page to +49 6227 8398 199) and please answer the following questions about

your current Oracle/SAP configurations.

Oracle Corporation

Oracle DB for SAP GTC

Altrottstr. 31

D-69190 Walldorf

Germany


4/82

Oracle database for SAP

is safe, reliable and scalable

C O N T E N T S1D A T A B A S E

3 Editorial

4 Oracle Database 11gRelease 2 for SAP10 Oracle Real Application Testing (RAT) Option certified for SAP

14 Oracle Real Application Testing (RAT) at Stadtwerke Munich (SWM)

16 SAP Partitioning Engine

18 Relationship Interview Oracle Database and SAP

20 Benchmark SAP on Oracle with Fujitsu

23 Benchmark SAP on Oracle Database 11gRelease 2 with Fujitsu

24 Benchmark SAP on Oracle-Solaris

28 SAP on Real Application Clusters (RAC) at Rpublique Franaise, Ministre de la Dfense

30 SAP on Real Application Clusters (RAC) at Southwest Airlines

34 Oracle Database Vault for SAP customers

36 SAP on Real Application Clusters (RAC) at Coca-Cola Hellenic

39 Oracle Index Compression at Tyson Foods

41 SAP on Real Application Clusters (RAC) at Sodexo Brazil

42 SAP on Real Application Clusters (RAC) at Verizon

44 SAP on Real Application Clusters (RAC) at Essent Netherlands

46 SAP on Oracle at Steelcase, HP-UX

50 Oracle Partner Connectria

52 SAP on Oracle Database at Dow Corning

54 SAP on Real Application Clusters (RAC) at Bauerfeind58 Hosting on Oracle at Freudenberg IT

61 Oracle Partner Delta Consulting

62 Oracle Partner All4IT Group

64 Oracle Data Guard Freudenberg ITj

70 Oracle Advanced Customer Services (ACS) for SAP

74 SAP Fast Start with Oracle and HP

77 Oracle Partner IS4IT, SAP on RAC

78 Useful links Oracle Database for SAP customers

79 Imprint


5/82

3Oracle Database for SAP Th e #1 Database safe, re l iable, scalable, optimized for SAP customers

D E A R S A P C U S T O M E R ,

For 21 years Oracle Corporation and SAP AG has maintained an ongoing commitment to

our joint customers and have long-standing reseller and customer support agreements in

place, which provides enhanced access to Oracle database technology and world class

customer support.

In this edition, we focus on Oracle Database 11gRelease 2, and the most important

technologies that benefit SAP customers in terms of return on investment, space savings

and performance.

Oracle Database 11gRelease 2 Enterprise Edition comes with a wide range of options to

extend the worlds #1 database to help grow your business and meet your users performance, security and availability

service level expectations.

The innovation continues with Oracle Database 11gRelease 2 to lower IT costs and deliver a higher quality of service by:

Consolidating business applications onto fast, reliable and scalable databases (Real Application Clusters RAC)

Maximizing availability and eliminating idle data center redundancy (RAC, Data Guard, online Patching)

Compressing data onto low cost storage partitions for faster performance (Oracle Advanced Compression)

Securely protecting information and enabling compliance (Oracle Advanced Compression, Oracle Database Vault)

Doubling DBA productivity and reducing the risk of change (Diagnostic Pack, integrated in BR* Tools and SAP DBA

Cockpit).

The Exadata V2 Sun Oracle Database Machine the worlds fastest for any type of database workload and the only

database machine that does online transaction processing (OLTP). Oracle Exadata Storage Software runs on each

storage server performing data-intensive processing that reduces the volume of data returned to database servers.

Smart Scans, Storage Indexes, Data Mining Model Scoring and Hybrid Columnar Compression are all components of

Oracle Exadata Storage Software that enable extreme performance.

If you would like to adopt Exadata V2 Sun Oracle Database Machine for your SAP Applications please see the Exadata

for SAP request form in this edition or on www.oracle.com/sap.

Read more about the benefits, why customers moved from single instance to Oracle Real Application Clusters for SAP

(Coca Cola Hellenic) and why Oracle database has the best cost of ownership (Tyson Food, Partitioning and Compression).

Compliance and security are getting more and more important today, please see the Oracle Database Vault for SAP article.

Please read more about the Oracle Advanced Customer Services (ACS) for SAP customers, including: database

migrations, Oracle-to-Oracle (O2O), performance tuning, health check and internal training workshops.

For more information or to see current and previous editions go to: www.oracle.com/sap.

We welcome your comments and questions, please contact us at: [email protected]

Sincerely

Gerhard KupplerSenior Director Corporate SAP Account,

Oracle Corporation


6/82

4

S A P C E R T I F I E D O R A C L E D A T A B A S E 1 1g

R E L E A S E 2 F U N C T I O N A L I T Y

I N T R O D U C T I O N

Oracle Database 11g Release 2 Enterprise Edition

delivers industry leading performance, scalability,

security and reliability on a choice of clustered or

single-servers, running Windows, Linux, and UNIX.

It provides comprehensive features to easily manage

the most demanding transaction processing, business

intelligence, and content management applications.

Oracle Database 11g Release 2 Enterprise Editioncomes with a wide range of options to extend the

worlds #1 database to help grow your business and

meet your users performance, security and availability

service level expectations.

SAP has certified Oracle Database 11g Release 2

(please see SAP Note 1398634) with the following

details:

Only SAP products based on SAP Kernel 6.40, 7.x

and higher are certified with Oracle Database 11.2.

The certification and rollout by SAP happens in

three waves: End of March 2010 (wave 1): major UNIX plat-

forms (AIX, HP-UX IA64, Solaris SPARC & x64)

and Linux x64 were released (done)

End of Q2/2010 (wave 2): Windows x64 and HP-

UX PA-RISC platforms are planned to be released

Windows IA64 tbd

Note:

A general release of Oracle Database 11.2 is not pro-

vided for earlier SAP releases (SAP R/3 3.1I up to and

including SAP R/3 4.6C). Similar to Oracle Database

Version 10.2, there will only be a temporary 11.2

release in direct connection with an SAP upgrade

project for these older SAP versions.

Oracle Database 11g Release 2 provides customers

more benefits: saving disk space with lower hardwarecosts, more performance, higher security, better

manageability, exceeding productivity and at least in

an outstanding high availability/disaster recovery for

SAP applications.

The following provides a list of important features

available for SAP customers today and many of them

are unique to Oracle.

Oracle Real Application Clusters 11gRelease 2

The SAP release of Oracle Database 11.2 RAC occurs

at the same time as the general release of OracleDatabase 11.2 in the SAP environment. For information

about Oracle RAC support, see SAP Note 527843.

It contains important details about the released RAC

configurations, like the fact that all used software stacks

with 10g RAC will continue to stay in place and can

be used with 11g RAC, so neither support for raw

devices (not supported any more with 11.2) nor ASM

is available for 11.2 RAC databases in SAP environ-

ments.

OLTP Table Compression

Oracle Database 11g introduced a new feature called

OLTP Table Compression that allows data to be

compressed during all types of data manipulation

operations, including conventional DML such as

INSERT and UPDATE. In addition, OLTP Table

Compression reduces the associated compression

overhead of write operations making it suitable for

transactional or OLTP environments as well. OLTP

Table Compression, therefore, extends the benefits

of compression to all application workloads.


7/82

Oracle Database 11g Release 2 for SAP 5

Oracles OLTP Table Compression, part of Advanced

Compression, uses a unique compression algorithmthat eliminates duplicate values within a database

block, even across multiple columns. Compressed

blocks contain a structure called a symbol table that

maintains compression metadata. When a block is

compressed, duplicate values are eliminated by first

adding a single copy of the duplicate value to the

symbol table. When compared with competing com-

pression algorithms that maintain a global database

symbol table, Oracles unique approach offers signifi-

cant performance benefits by not introducing addi-

tional I/O when accessing compressed data.

The benefits of OLTP Table Compression go beyond

just on-disk storage savings. One significant advantageis Oracles ability to read compressed blocks directly

without having to first uncompress the block.

Therefore, there is no measurable performance degra-

dation for accessing compressed data. In fact, in many

cases performance may improve due to the reduction

in I/O since Oracle will have to access fewer blocks.

Further, the buffer cache will become more efficient

by storing more data without having to add memory.

Since Oracle 10.2.0.2, SAP has certified the use of

Index compression to save disk space for indexes and

reduce total database size on disk. Customer experi-

ences show that even after a full database reorganiza-

tion has taken place, an additional 20% of total diskspace reduction for the whole database can be achieved

using index compression.

SecureFile Data Compression

SecureFiles is a new feature in Oracle Database 11g

that introduces a completely re-engineered large object

(LOB) data type to dramatically improve performance,

manageability, and ease of application development.

SecureFiles data is compressed using industry standard

compression algorithms. Compression not only results

in significant savings in storage but also improved

performance by reducing IO, buffer cache requirements,redo generation and encryption overhead. SecureFile

compression provides significant storage and handles

in-line and out-of-line LOB data which are getting

more and more important in SAP applications and are

widely used in SAP products such as SAP CRM, SAP

XI, SAP NetWeaver Portal, and even in SAP ERP.

Almost all non-cluster tables in SAP ERP use out-of-

line LOBs that are unique to the Oracle database.

RMAN Backup Compression

The continuous growth in enterprise databases createsan enormous challenge to database administrators.

The storage requirements for maintaining database

backups and the performance of the backup procedures

are directly impacted by database size. Oracle Advan-

ced Compression includes compression technology

that can dramatically reduce the storage requirements

for backup data. Due to RMANs tight integration

with Oracle Database, RMAN backup (widely used

by SAP customers) data is compressed before it is

written to disk or tape and doesnt need to be uncom-

pressed before recovery providing an enormous

reduction in storage costs.

Data Pump Compression

The ability to compress the metadata associated with

a Data Pump job was first provided in Oracle Data-

base 10g Release 2. In Oracle Database 11g, this

compression capability has been extended so that

table data can be compressed on export. Data Pump

compression is an inline operation, so the reduced

dump file size means a significant savings in disk

space. Unlike operating system or file system com-

pression utilities, Data Pump compression is fully

inline on the import side as well, so there is no need

to uncompress a dump file before importing it. Thecompressed dump file sets are automatically decom-

pressed during import without any additional steps

by the Database Administrator.

Data Guard Network Compression

Data Guard provides the management, monitoring,

and automation software infrastructure to create,

maintain, and monitor one or more standby databa-

ses to protect enterprise data from failures, disasters,

errors, and data corruptions. Data Guard maintains

synchronization of primary and standby databases

using redo data (the information required to recovera transaction). As transactions occur in the primary

database, redo data is generated and written to the

local redo log files. Data Guard Redo Transport

Services are used to transfer this redo data to the

standby site(s).

11gD A T A B A S E


8/82

6

With Advanced Compression, redo data may be

transmitted in a compressed format to reduce thenetwork bandwith in an efficient way. Redo transport

compression is no longer limited to compressing

redo data only when a redo gap is being resolved.

When compression is enabled for a destination, all

redo data sent to that destination is compressed.

Deferred Segment Creation

Beginning in Oracle Database 11g Release 2, when

creating a table in a locally managed tablespace,

table segment creation is deferred until the first row

is inserted. In addition, creation of segments is deferred

for any LOB columns of the table, any indexes created

implicitly as part of table creation, and any indexes

subsequently explicitly created on the table.

The advantages of this space allocation method for

customers running Oracle database underneath their

SAP applications are the following:

Empty database objects will not consume any

disk space

Very important for SAP environments as 60-70%

of all tables, lobs, indexes and partitions in an SAP

installation are empty

Makes database installation for SAP a lot faster

because creation of empty tables, LOBs, and indexes

are dramatically faster.

Oracle Data Dictionary Space queries run

substantially faster

Enhanced Add COLUMN Functionality

Before Oracle 11g adding new columns with

DEFAULT values and NOT NULL constraint required

both an exclusive lock on the table and the default

value to be stored in all existing records.

Now in Oracle 11g the database can optimize the

resource usage and storage requirements for thisoperation, default values of columns are maintained

in the data dictionary for columns specified as NOT

NULL. Adding new columns with DEFAULT values

and NOT NULL constraint no longer requires the

default value to be stored in all existing records.

This not only enables a schema modification in sub-

seconds and independent of the existing data volume,

it also consumes no space. Especially for large tables,

updating table column results in reduced execution

time and space saving.

Because add column is very common within SAPBW applications and SAP upgrades, enhanced ADD

Column Functionality leads to:

factor 10-20 performance improvement for SAP

BW during add column process

saving large amount of disk space

Bitmap Indexes

Especially SAP NetWeaver Business Warehouse

customers now can use Bitmap Indexes. Bitmap

Indexes are high densed indexes which need minimal

disk space. This reduces the size of a database very

effectively. Oracle is the only database vendor whooffers disk-based Bitmap Indexes.

Variable Length Datatype Representation

Also Oracle is the only vendor to provide a compre-

hensive Variable Length Datatype Representation.

With Variable Length Datatype Representation

minimum disk space is allocated. Bitmap Indexes

and Variable Length Datatype Representation are the

main reasons why the Oracle database is smaller than

databases from other vendors.

Compared to Oracle Database 10g the listed OracleDatabase 11g features give the following results:

First the size of the 11.2 database will be 3 times

smaller. So a 10.2 database with a size of 3 Terabytes

will only consume 1 TB in disk space with 11.2.

Second backups will also be smaller up to 10 times

smaller.

11gD A T A B A S E


9/82

7

SecureFile Performance

SecureFiles offer the best solution for storing filecontent, such as images, audio, video, PDFs, and

spreadsheets. Traditionally, relational data is stored

in a database, while unstructured contentboth

semi-structured and unstructuredis stored as files

in file systems. SecureFiles is a major paradigm shift

in the choice of files storage. SecureFiles is specifically

engineered to deliver high performance for file data

comparable to that of traditional file systems, while

retaining the advantages of the Oracle Database.

SecureFiles offers the best database and file system

architecture attributes for storing unstructured content.

SAP customers benefit from SecureFiles because of

Significantly faster access times compared to LOBs

in SAP environments

Increased transaction throughput on SAP cluster

tables especially with RAC

Prerequisite for compression of SAP tables

containing LOBs (e.g. cluster tables)

Overall transaction throughput increases when LOB

data is stored in SecureFiles (see figure below). LOB

data stored in SecureFiles delivers equal or better

performance compared with LOB data stored in

LONG or BasicFiles (LOB implementation prior to

11g). SecureFiles improve dramatically the scalability

of SAP applications running against Oracle Database

11g RAC but also Oracle Database 11g Single Instance

benefits substantially from SecureFiles. Therefore a

clear recommendation is given to migrate all existing

LONG and Basicfile LOB data to SecureFiles.Hi

Tablespace Encryption

On many agendas security is in the forefront and for

customers becoming much more important. Oracle

has a lot of features in the area of security. Especially

for SAP customers it is now very easy to encrypt sen-sitive data. Included within the Advanced Encryption

is the technology to encrypt all data in a tablespace.

Prior to 11g, Oracle provided security features such

as Column Encryption through Transparent Data

Encryption (TDE) and Client Server (SAP Application

Server) Network Encryption.g

Additional Encryption

Advanced Encryption within 11g includes RMAN

Backup Encryption, so no one can read data from a

(stolen) backup. In addition customers can easily

encrypt export file with Oracle Data Pump unstruc-

tured LOB data with SecureFile Encryption.

DG Secure Network Transport

Data Guard provides a secure environment and

prevents tampering with redo data as it is being

transferred to the standby database. To enable secure

transmission of redo data, set up every database inthe Data Guard configuration to use a password file,

and set the password for the SYS user identically on

every system.

More Security Database Vault

Oracle Database Vault is also certified for use with

SAP applications. With Oracle Database Vault, pro-

tective realms around SAP application database objects

can be established to prevent privileged database users

from accessing sensitive data and to enforce separation

of duties among privileged database users.

row size in KB

LONGs (9.2,10.2) LOBs (10.2,11.2) Securefiles 11.2

Performance

ImprovementFactor

SAP VBDATA Throughput (Insert/Read/Delete)

Oracle Database 11g Release 2 for SAP


10/82

8

Real Application Testing (RAT)

Database Replay provides DBAs and system admi-

nistrators with the ability to faithfully, accurately

and realistically rerun actual production work-

loads, including online user and batch workloads,

in test environments. By capturing the full data-

base workload from production systems, including

all concurrency, dependencies and timing, Database

Replay enables you to realistically test system

changes by essentially recreating production work-

loads on the test system, something that a set of

scripts can never duplicate. With Database Replay,

DBAs and system administrators can test:

Database upgrades, patches, parameter, schemachanges, etc.

Configuration changes such as conversion from

a single instance to RAC, ASM, etc.

Storage, network, interconnect changes

Operating system, hardware migrations,

patches, upgrades, parameter changes

SQL Performance Analyzer (SPA) can predict and

prevent SQL execution performance problems

caused by environment changes. SQL Performance

Analyzer provides a granular view of the impact of

environment changes on SQL execution plans and

statistics by running the SQL statements seriallybefore and after the changes

Direct NFS

With Oracle Database 11g Release 2, you can configure

Oracle Database to access NAS devices directly using

Oracle Direct NFS Client, rather than using the

operating system kernel NFS client. Oracle Database

will access files stored on the NFS server directly

through the integrated Direct NFS Client eliminating

the overhead imposed by the operating system kernel

NFS. These files are also accessible via the operating

system kernel NFS client thereby allowing seamlessadministration.

SAP customers can benefit from Direct NFS in the

following way: Improves throughput of NAS solutions such as

NetApp

Up to 50% more database throughput in NAS

environments with multiple NICs

Up to 20% CPU savings on database server

Works for Single Instance and Real Application

Clusters (RAC)

Works for UNIX/Linux and Windows Platforms

Highly Available Network Solution

Faster, easier and more available than any OS or

NAS based bonding or trunking solution Direct NFS with NAS may provide higher

throughput than traditional, more complex SAN

solutions

Superior to any bonding solution faster and

easier

Better throughput than most SAN solutions

Online Patching

A regular RDBMS patch is comprised of one or more

object files and/or libraries. Installing a regular patch

requires shutting down the RDBMS instance, re-linking the oracle binary, and restarting the instance;

uninstalling a regular patch requires the same steps.

With Oracle Database 11g, it is possible to install

single or bundle patches completely online, without

requiring the database instance to be shut down, and

without requiring RAC or Data Guard configurations.

With online patching, which is integrated with

OPatch, each process associated with the instance

checks for patched code at a safe execution point,

and then copies the code into its process space.

Snapshot Standby

Easy conversion of a physical standby database to a

reporting database A physical standby database can

be opened read/write for reporting purposes, and then

flashed back to a point in the past to be easily conver-

ted back to a physical standby database. At this point,

Data Guard automatically synchronizes the standby

database with the primary database. This allows

the physical standby database to be utilized for

read/write reporting activities for SAP applications

e.g. NetWeaver BI.

11gD A T A B A S E


11/82

oracle.com/[email protected]

Only Oracle Database 11gRelease2 provides for

SAPcustomers

Copyright 2010, Oracle. All rights reserved. Oracle is a registered trademark of Oracle Corporation and/or its a affiliates.

Other names may be trademarks of their respective owners.

Oracle the # 1 Database for Deploying SAP Applications

Oracle other DBsBest Performance & Scalability

Fully integrated, platform independent active-active clustering solutionfor both SAP and database through RAC and Clusterware/SAPCTL

Dictionary only add column with default High Performance NAS Storage through Direct NFS

Best Deployment Flexibility

Online Patching Real Application Testing Option

Best Availability & Reliability Updatable Standby Database Fully integrated, platform independent active-active clustering solution

for both SAP and database through RAC and Clusterware/SAPCTL

Best Support for Very Large Databases

RMAN Backup Compression SecureFile out-of-line LOB data compression Deferred Segments Creation Disk-Based Bitmap Indexes Dictionary only add column with default High Performance NAS Storage through Direct NFS

Best Database Security

Tablespace Encryption

RMAN Backup Encryption

Data Guard Secure Network Transport

Database Vault

Best Manageability & Self-Management

Real Application Testing Option

Online Patching

Automatic Workload Repository (AWR)


12/82

10

R E A L A P P L I C A T I O N T E S T I N G C E R T I F I E D F O R S A P

Making a change to the SAP database can represent

a significant risk to an organization. The effects of an

unexpected problem could be widely felt and have

serious consequences. For this reason, testing of chan-

ges prior to production deployment is considered

critical. The cost involved in testing using traditional

methods can be very high both in time and money.

Even after testing, many problems will only be dis-

covered in production, and as a result end-users will

suffer from poor performance or possibly even outages.

Oracle now has a new and unique testing tool for the

Oracle Database called Real Application Testing.

This Database option is comprised of two complemen-

tary features, SQL Performance Analyzer (SPA) and

Database Replay. Together, these features can help

ensure that database changes do not negatively impact

SAP availability and performance. They significantly

improve the quality and lower the cost of testing

changes to the SAP database system.

SQL Performance Analyzer (SPA)

One of the key tasks for a DBA is to tune SQL state-

ments for optimal performance. Changes to databasesystems (such as optimizer statistics refresh, database

upgrades, and patches) can cause the performance of

previously well tuned SQL statements to regress, which

can dramatically impact system performance for end-

users. Until now, it was difficult if not impossible to

predict how day to day operational changes to the

database will impact application SQL performance.

With SQL Performance Analyzer, or SPA, you can

now know exactly how SQL will behave after routine

system changes.

SPA automatically captures production SQL state-

ments from your SAP database and re-executes them

following a change to measure the impact. The state-

ments are executed serially, in isolation from each

other and regressed statements are automatically hig-

hlighted. Integration with SQL Plan Management

and SQL Tuning Advisor enables automatic tuning

of these statements to address any regressions prior

to a change being implemented in production.

Orchestration of the entire SPA process can be done

using Enterprise Manager, which provides a number

of workflows for testing common operations such as

Database Upgrades. There is also a Guided Workflowwhich enables custom experiments to be conducted.


13/82

11

Real Application Testing was introduced in the

11g release of the Oracle Database, however, SQLPerformance Analyzer is also available for previous

releases of the Oracle Database. Consequently, SQL

Performance Analyzer can be used to test changes in

SAP environments even if there are no immediate

plans to move to Oracle Database 11g Release 2.

Some examples of changes that could be assessed

using SQL Performance Analyzer in SAP environ-

ments are:

Upgrade SAP Database from Oracle Database 10g

Release 2 to 11g Release 2

Install Database Patchset

Migration from single instance to RAC

Implement Partitioning

Implement Advanced Compression

Changes to FIX_CONTROLS parameter settings

Creation of Histograms on SAP tables

Some of these changes can actually be tested safely

on the production system, therefore removing the

need for establishing a separate test environment.

There are no SAP specific patches required to use

SQL Performance Analyzer. Further general informa-

tion on using SQL Performance Analyzer can be found

in the following documents:

Technical White Paper: SQL Performance Analyzer

Viewlet: SQL Performance Analyzer in Oracle Data-

base 11g

Oracle Database Real Application Testing Users Guide

Database Replay

Load testing today is generally done using tools that

allow testing teams to generate synthetic workloads

based on what they expect users to do on a system.These workloads can then be replayed by application

virtual users, which simulate the end users by sub-

mitting requests to the application. Although widely

used, this approach has a number of shortcomings

when it comes to testing database level changes:

Creating the synthetic workload can take a

considerable time and requires programmingexpertise

User behaviour is not well understood, so many

possible workflows are often missed in the synthetic

tests

Production scale database concurrency is near

impossible to simulate with these tools

A full application stack is required for testing as

these tools simulate end users

Database Replay makes it possible to capture a

workload on a production system with negligible

performance overhead and replay it on a test system

with the exact timing, concurrency, and transaction

characteristics of the original workload. Database

Replay is suitable for testing changes to the database

tier and anything below it, such as storage or OS

platform. This also means that tiers such as the

Web and Application Server tier are not required in

the test environment. The Database Replay capture

process has been backported and certified for SAP

databases running on Oracle Database 10g Release 2

(10.2.0.4).

Changes that Database Replay could be used to

validate include: SAP Database upgrade from 10gR2 (10.2.0.4)

to 11gR2

Platform migration such as from Windows to

Oracle Solaris

Database Patchset install

Changes to FIX_CONTROLS parameter settings

Schema changes

Migration from single instance to RAC

Implement Partitioning

Implement Advanced Compression

Database storage changes (e.g. Direct NFS)

Tablespace encryption

Object Reorganization

Oracle Real Appl ication Testing (RAT) Option cert i f ied for SAP


14/82

12

Database Replay is included with Oracle Database

11g. In addition, Database Replay capture can be runagainst Oracle Database 10.2.0.4 and higher version.

To use Database Replay in SAP environments, Patch

9233285 is required on the Replay database.

For more technical information on using Database

Replay please refer to the following whitepaper:

Technical White Paper: Database Replay

Viewlet: Database Replay in Oracle Database 11g

Oracle Database Real Application Testing Users Guide

Results of Testing

The Oracle/SAP Development Team recently conclu-ded comprehensive testing of SAP applications using

both SQL Performance Analyzer and Database Replay.

The testing consisted of SAP ERP and BW workloads

running in-house on the ABAP stack. Further testing

was undertaken at a large utilities customer running

the SAP IS-U (Industry Specific Utilities) module.

The objective was to use Real Application Testing

to measure the effects of Advanced Compression for

OLTP and SecureFiles on DML operations for SAP

ERP. The testing activity consisted of using Database

Replay to capture a production workload from a

10.2.0.4 SAP database. This workload was thenreplayed multiple times against both compressed

and non-compressed Oracle Database 11g Release 2

databases. Key AWR metrics from the replays are

shown below.

Load Profile

Redo size/s

Logical reads/s

Block changes/s

Physical reads/s

Physical writes/s

User calls/s

Parses/s

Hard parses/s

W/A MB processed/s

Executes/s

Transactions/s

Instance Efficiency Percentages (Target 100%)Buffer Hit %

Library Hit %

Host CPU (CPUs: 8 Cores: 8 Sockets: 4)

%User

%System

%WIO

%Idle

1)

153456,80

18748,80

459,70

3820,60

74,60

1806,50

247,20

32,70

12,50

2572,10

4,50

81,91

95,08

12,30

4,40

12,30

81,80

203407,00

18210,00

595,20

1635,80

72,00

2042,10

254,10

33,60

12,70

2834,60

4,80

91,79

95,52

12,80

4,20

11,20

81,80

32,55

-2,87

29,48

-57,18

-3,49

13,04

2,79

2,75

1,60

10,21

6,67

12,06

0,46

4,07

-4,55

-8,94

0,00

DB size after normal reorg: 3562,00 GBDB size after reorg and OLTP + SecureFiles (med) + Index compression: 1589,00 GB

Space saved (~ 55%) 1973,00 GB

KEY AWR Metrics

Replay without

Compression

Replay with

Compression Diff%

1) normalized redo (+6,67%) is 163692/s for uncompressed run


15/82

Oracle Real Appl ication Testing (RAT) Option cert i f ied for SAP 13

In order to further analyze the effects of Advanced

Compression on DML operations, SQL PerformanceAnalyzer (SPA) was used concurrently with Database

Replay. As the workload was being replayed,

SPAs automatic cursor cache capture feature was

used to capture the SQL statements into a SQL

Tuning Set. SQL Tuning Set contains the text of the

SQL statements along with their bind values,

execution plans and statistics. This was done for both

replays, with compression enabled and without

compression. In all over 5000 SQL statements were

analyzed. The table below shows the SQL Tuning

Set comparison of some the most frequently executed

DML operations in the SAP workload.

Overall the test results were very instructive. Key

findings are described as follows (note: results will

vary according to your workload):

No significant overhead observed during capture

process

Advanced Compression reduced database size by 50%

Redo generated increased by ~25%

Physical reads reduced by 60%

CPU usage stayed flat

Problems encountered:

During the testing bug # 9233285 was encountered.Patch for this bug must be applied before using

Real Application Testing with SAP systems.

Conclusion

Real Application Testing proved a vital tool forvalidating upgrade from Oracle Database 10g

Release 2 to 11gRelease 2. The ability to test with

production workloads and SQL statements is

essential for testing business critical applications

like SAP and this feature will significantly mitigate

upgrade and change risk when used by experienced

DBAs.

Sample SQL Tuning Set Comparison

Compression SQL ID SQL Text Executions ElapsedTime

BufferGets

DiskReads

ModuleCPUTime

N g9makkh1cnq94 INSERT INTO ALCLASTOOL VALUES( :A0 , :A1 , :A2 , :A3 , :A4.. . 37648 180,29 152,97 1136194 32901 SAPLSALU

Y g9makkh1cnq94 INSERT INTO ALCLASTOOL VALUES( :A0 , :A1 , :A2 , :A3 , :A4.. . 40434 215,54 164,10 1118975 33911 SAPLSALU

N 6kxdz6nknay0 m INSERT INTO ALGRPCUSGE VALUES( :A0 , :A1 ,:A2 , :A3 , :A4... 18257 80,88 71,32 518407 9978 SAPLSALU

Y 6kxdz6nknay0m INSERT INTO ALGRPCUSGE VALUES( :A0 , :A1 , :A2 , :A3 , :A4.. . 18889 92,12 74,99 532217 9615 SAPLSALU

N c42810chs804v INSERT INTO ALGRPCUSPF VALUES( :A0 , :A1 , :A2 , :A3 , :A4... 4100 20,51 16,67 134523 3652 SAPLSALU

Y c42810chs804v INSERT INTO ALGRPCUSPF VALUES( :A0 , :A1 , :A2 , :A3 , :A4... 4202 24,07 17,32 137029 3493 SAPLSALU

N 8wz2ywwcrh7c 7 INSERT INTO ALGRPCUSMC VALUES( :A0 , :A1 , :A2 , :A3 , :A4... 2144 12,48 9,08 77571 2758 SAPLSALU

Y 8wz2ywwcrh7c 7 INSERT INTO ALGRPCUSMC VALUES( :A0 , :A1 , :A2 , :A3 , :A4... 1820 13,18 7,94 65849 2289 SAPLSALU


16/82

14

Making a change to the SAP database can

represent a significant risk to an organization.

The effects of an unexpected problem could

be widely felt and have serious consequences.

For this reason, testing of changes prior to

production deployment is considered critical.

The cost involved in testing using traditional

methods can be very high both in time and

money. Even after testing, many problems will

only be discovered in production, and as a

result end-users will suffer from poor perfor-

mance or possibly even outages.

The Real Application Testing (RAT) Option

is comprised of two complementary features,

SQL Performance Analyzer (SPA) and Database

Replay. Together, these features can help ensure

that database changes do not negatively impact

SAP availability and performance. They signi-

ficantly improve the quality and lower the cost

of testing changes to the SAP database system.

Facts instead of surprises

The IT specialists at SWM, are among the

pioneers who are using the Oracle Database 11gRelease 2 Real Application Testing option

together with the SAP application.

SWM Stadtwerke Mnchen, a public utility

and service company based in the mega city of

Munich that offers a wide range of products,

has data volumes of more than 3 Terabyte

which are being accessed by 1500 to 1800

SAP users. The CPU utilization of the pro-

duction systems is therefore extremely high.

For Christian Duschl, SAP Team-Leader System

Operations Management, the risks and con-

tingencies associated with making changes to

the SAP environment were always a high-cost

challenge: Previously, we were not able to

determine the dialog load with sufficient

precision before commissioning new systems.

Surprises were part of our day-to-day work.

Having used RAT for the first time and gai-

ned some experience with it, we believe that

this will finally become a thing of the past.

The test scenario was compressing the databaseusing the Advanced Compression option and

to examine the effects on system performance.

The production load of two RAC nodes was

recorded over two working days and the data-

base was transferred to a test server along with

the data volume for this period.

In a first step, the database was reorganized in

a conventional manner, without being com-

pressed. A representative 12 hour section of the

recorded production load was then replayed.

The replay contained the SAP ERP user load

only. Other applications, such as those from the

call center area (non-SAP), were not taken into

account and were filtered out during replay.

The results from the replay were later used as

a basis for further analysis. In the next step,

all tables and indexes were compressed using

OLTP Table, SecureFiles and Index compres-

sion.

Stadtwerke Munich

Services GmbH

Industry:

Utility/Public Sector

Annual Revenue:

4.2 billion

Employees:

Around 6,800

Im

S U C C E S S F U L T E S T O F O R A C L E D A T A B A S E 1 1 gR E L E A S E 2 R E A L A P P L I C A T I O N T E S T I N G O P T I O NA T S TA D T W E R K E M U N I C H ( S W M )

The positive results from using the RAT option for the first time enable us to look to the

future and the upcoming changes in a much calmer manner than before.

C H R I S T I A N D U S C H L ,

Teamleader System Operations Management


17/82

Oracle Real Appl ication Testing (RAT) at Stadtwerke Munich (SWM) 15

A second run, which included exactly the same

dataset as the first run, was then tested againstthe compressed database and the results were

compared with the first run.

No additional costs and less disk space

Manfred Fischer, Manager of SAP system

maintenance in information and process tech-

nology and general coordinator of the RAT

project, summarizes the results as follows:

When we were recording the basic data over

two days for the test, we couldnt find any

negative effects on the CPU or I/O load of theproduction system and the dialog response

times didnt change.

The compression result was quite impressive.

The amount of disk space shrunk from almost

3.6 TB to around 1.6 TB, resulting in a space

saving of 55%.

What was even more interesting was the

effect on the CPU load when using com-

pression. The result was amazing. Thanks

to the enormous reduction in I/O of

almost 60% and considerably better uti-lization of the main memory, the additio-

nal costs expected for compression were

compensated in full for the CPU load.

Another positive side-effect was a slight

increase in transaction throughput.

Because the hit rate of the database buffer

cache improved considerably, there was

less physical I/O to the disk subsystem.

As the solution does not show any performance

impacts, the Advanced Compression option can

be used on the SAP ERP production systems

at Stadtwerke Mnchen if necessary.

According to Manfred Fischer, SQL Perfor-

mance Analyzer (SPA) also proved to be a key

functionality. It was particularly helpful in

evaluating the performance of important

queries at statement level, which must not be

negatively impacted.

Calculated risks and cost savings

In the future, the SAP environment at Stadt-

werke Mnchen is to be split into three units

(Network, Delivery and Coresystem) which

will be designed to mirror the data volumes

in the production system and various test and

security systems, among other things.

Because each one of these segments should

have the capacity of the current system, the

data volumes will triple in the final expansionstage.

Given such complex tasks, Christian Duschl

is giving an extremely positive conclusion

from the test results with RAT:

The positive results from using the RAT

option for the first time enable us to look

to the future and the upcoming changes

in a much calmer manner than before.

Unknown risks when making release changes,

growing data volumes or deregulating an

entire SAP system, as was the case here, loosetheir fears for many users at least for the data-

base level. Using RAT will enable IT depart-

ments to demonstrate extremely positive

cost-value results to the controlling depart-

ments.


18/82

16

S A P P A R T I T I O N I N G E N G I N E R E L E A S E D B Y S A P

The SAP Partitioning Engine provides a solution for SAP on Oracle database systems that can use a

predefined set of about 30 large application tables for partitioning, based on a technology concept from

Oracle. SAP Partitioning Engine was released for customers in January 2010. The solution offers full

integration of range partitioning into SAP OLTP products (e.g. ERP).

Unlike SAP BI, partitioning is not used by default

in the SAP ERP system, for mainly historical reasons.

Today, SAP installations typically have database sizes

of around 1TB, whereas SAP customers using special

functionalities such as the SAP retail module are

running databases of 10TB and greater.

Such large databases have several challenges:

Old data must be archived, but archiving can have

a negative impact on database performance.

Deleting data from the database generates a huge

transaction load, affecting normal data processing.

Archiving very large tables can be a never-ending

job. Once archiving is complete, follow-up tasks

are needed either to rebuild fragmented indexes

and/or to free up space in the tablespaces. These

additional tasks occupy maintenance windows for

the system, reducing overall availability. Response times to user queries and batch processing

must be guaranteed. This becomes more and more

difficult because the tables are constantly growing

in size.

Processing larger tables uses up much more time

and resources on the database as well as on the

application servers. Because of the longer runtime,

work processes take longer which means that more

powerful application servers are needed.

Fully integrated

The principle of partitioning is splitting a large

table in smaller parts, called partitions, grouping

and separating the rows of the table based on the

content of one or more columns. Rows fulfilling

the same criteria are placed in the same partition.This also allows queries of a table to execute faster

because the data will now be grouped, instead of

being distributed across the whole table.

Complex analysis and maintenance jobs of partitioning

can be automated. Therefore, Oracle in the initial

phase of the project had developed a set of tools

that was based on Oracles own technologies, mainly

PL/SQL procedures. This worked fine for a few

installations but, it was a consulting based solution

which is useful for non-standard SAP requirements,

it should not be necessary for standard environments.

To make the innovative method available to morecustomers SAP has developed their own solution,

based on the concept provided by Oracle.

The SAP Partitioning Engine is fully integrated in

the SAP system. It offers a predefined set of about

30 application tables to be partitioned based on time

based criteria. So existing non-partitioned tables will

be converted through ABAP/SAP brtools task. Also

partition maintanance is fully automated through

the internal SAP SM37 job, requiring no DBA inter-

vention.

A partitioning service for non-BI SAP systems has beenavailable from Oracle for several years but, until now

was not integrated into SAPs DB admin tools. The

Partitioning Engine will be available with WEB AS

6.20, SAP ECC 5.0/6.0, Business Suite 7, brtools 7.10

and running Oracle 10.2.0.4 or higher. For other

prerequisits please see SAP Note 1333328. For part-

itioning without the Partitioning Engine, please see

SAP Note Oracle Partitioning 742243 and SAP note

722188 FAQ.

Dr. Christian Graf, Development Manager, DB Platforms Oracle & Informix,

OS Platforms HPUX, Solaris, Tru64 SAP AG


19/82

SAP Part it ioning Engine 17

The bulding concept of the SAP Partitioning Engine

is based on the following three main parts:

Analysis and Statement Creation

First part is an ABAP program, which analyzes the

content of the document column of an application

table and creates as output a DDL (Data Definition

Language) statement for the table and indexes with

suitable partition ranges. This kind of transaction

initially generates non-partitioned tables into parti-

tioned ones. This includes built-in functionality.

Such as well-known SAP tables, their partitioning

key as well as an automatic adjustment for partition

sizes. Analysis task can be done even on very large

tables in short time with the help of the underlaying

number-ranges object defined in table NRIV

(NRIV Partitioning).

Partition Transformation

Second part of partitioning engine is a new function

in the brspace tool, which reads the previously created

DDL statement and performs the online redefinition.

It includes LOB migration if the table contains long

field. The new online table generation can be done

without downtime, because brspace use Oracle

Technology (DBMS_REDEFINITION-Package).

Partitioning Care Process

Third part is also an ABAP program, specifically for

maintaining the partitioning schema. One maintaining

task done by this program is that if new transaction

data are inserted in the partitioned table automatically

additional partitions are created if needed. If the table

is archived, the number of rows stored in the oldest

partitions will decrease. The transaction of the SAP

Partitioning Engine is responsible for freeing up the

space no longer used in these partitions. The space

is freed by executing a merge operation on the old

partitions. The result is one small partition containingthe remaining rows and a lot of free space within the

tablespace, which can be used for creating new parti-

tions. This transaction can also be executed while

SAP is up and running.

Advantages and Benefits

The SAP Partitioning Engine avoids negative side-effects when archiving by seperating historic and

current application data using range partitioning

(Oracle offers several other types of partitioning

including: list and hash partitioning, which are

normally not suitable for SAP ERP systems). Range

Partitioning has a couple of advantages or benefits:

Index and table formatting is restricted to the

partition with the data records to be archieved.

Archiving transactions and running application

transactions access different partitions of the table.

Block usage is improved by avoiding mixing old

data and new data.

Further advantage, disk space is reclaimed by

merging the partitions that are empty after

archiving and partitioning can also improve

query performance.

Using the SAP Partitioning Engine allows SAP

customers to make use of the Oracle partitioning

technology, without the need to be familiar with

Oracle programming. The predefinition of the most

common candidate tables for partitioning allows less

experienced Oracle users to benefit from this innova-

tive approach.

Starting Partitioning to select a table with SAP Partitioning

Engine (Source: SAP)


20/82

18

1.) Mr. Schulte,

the database partnership with Oracle has

thrived for more than 20 years. Can you

summarize this for us in a few key words?

Schulte:

Firstly, our intent is to continue to satisfy the tens of

thousands mutual SAP on Oracle database customers.

Our joint work has always been characterized by a

constant desire to provide mutual customers withefficient service and support solutions for their appli-

cation needs in order to bring additional benefit to

their business and to offer optimum protection of

their investments.

Secondly, continually accelerating and optimizing

the use of safe, reliable and scalable database techno-

logy in a mission-critical application environment,

something we have achieved time and again.

And thirdly, we have enjoyed a very successful service

and support partnership with Oracle Corporation.

For the benefit of our mutual customers, we have along term contractual frame work and resources in

place driving this forward. The Oracle team suppor-

ting our joint relationship has been in place for more

than 10 years, this stability in personnel has been very

benefical to both SAP and our mutual customers.

The Oracle SAP resources include database develop-

ment engineers and 24 by 7 follow the sun customer

support.

R E L A T I O N S H I P I N T E R V I E W O R A C L E D A T A B A S E A N D S A P

Horst Udo Schulte, Vice President Global Partner Programs, SAP AG

Gerhard Kuppler, Senior Director Corporate SAP Account, Oracle Corporation

Listen to the Oracle webcast at www.oracle.com/sap

2.) Mr. Kuppler,

what importance does Oracle attach to the

database technology partnership with SAP?

Kuppler:

Satisfied mutual customers !

For more than 20 years Oracle has been the safe,

reliable, and scalable database of choice for SAP

customers.

The Oracle database is constantly optimized for SAP

Business Suite Applications and around the world

more than two thirds of all mid-size to high-end SAP

customers in every industry, entrust their application

deployments to Oracle databases.

Customers are running SAP applications with Oracle

databases on the same code base on Unix, Linux andWindows operating systems.

Oracle was one of the technology partners involved

in developing SAP R/3, when SAP began developing

R/3 in 1988. Oracles database technologies and

SAPs solutions are very closely linked and perfectly

integrated with one another.

We managed to establish our Oracle database techno-

logies as the number one database amongst SAP

customers around the globe.

The very large and growing customer base expects acost benefit from our two companies technologies in

the long term and also to receive first-class database

support from both of us.

Oracle is deploying a significant amount of resources

at SAP HQ Walldorf Germany, Palo Alto and Tokyo

to deliver safe, reliable and scalable DB technology

and round the clock services with a great long term

commitment to the mutual customer base.

Gerhard Kuppler Horst Udo Schulte


21/82

19Relationship Interview Oracle Database and SAP

3.) Mr. Schulte, Mr. Kuppler,

for you what are the main objectives for both

parties continuing to successfully drive this

relationship forwards in the future?

Schulte:

We are committed to the success of our mutual

customers.Since 1988, SAP AG has been using Oracle as a Tier 1

development database, on which SAP develops, ports

and migrates its products to function optimally with

Oracles database technologies.

This will once again be underlined with the new

Oracle Database 11g Release 2. It will contain various

specific functionality designed and implemented for

the needs of the SAP Business Suite.

Kuppler:

Let me add, the release strategy of SAP AG for Oracle

Database versions is to certify only the terminalrelease of a new database version. This strategy is

based on the overall requests of the SAP customer

community to minimize the number of version

upgrades in complex SAP environments. It has been

proven a successful roll out process with previous

terminal releases.

We are committed to maintain a productive Database

relationship with SAP to keep the mutual base and

also new SAP on Oracle customers satified and

provide them with the latest Database innovations

like: Real Applications Clusters, Database Vault and

new Oracle Database 11g functionality.

Resulting in scalable performance, cost savings,

better service levels and easy system administration.

4.) Mr. Schulte, Mr. Kuppler,

how frequently does SAP and Oracle renew their

partnership agreements?

Schulte:

We have renewed our agreement approximately

every 3 years and we plan to continue this process

for the benefit of our joint customers.

Kuppler:With 2/3 of all mid-size to high-end SAP customers

running on Oracle databases we have a mutual

interest in doing what is best for our joint install

base.


22/82

20

In October and December 2009 Oracle and

Fujitsu strikingly demonstrated the success of

their longstanding partnership with joint press

announcements about two world record results

in a row with the SAP Business Intelligence

Data Mart Standard Application Benchmark.

The SAP BI Data Mart Standard Application

Benchmark provides a real-world indication of

enterprise database performance by measuring

system performance in handling high-volume

queries to a business warehouse system data-

base. The key figure is the number of query

navigation steps per hour against an enormous

amount of data.

The first announcement was made during

Oracle Open World about a two-tier world

record result achieved on a system comprised

of a single-node Fujitsu PRIMERGY RX300

S5 server(1)

, Oracle Database and Oracle RealApplication Clusters on Linux. The test was also

executed on a two-node Fujitsu PRIMERGY

RX300 S5 cluster. The one-node result (320,

363 navigation steps per hour) and two-node

result (2) (609,349 navigation steps per hour) not

only indicated excellent scalability, but also

delivered more than 67% higher performance

per core and 3.3 times more performance-per-

processor than the highest IBM DB2 results (3)

on the SAP BI-D Standard Application Bench-

mark.

Performance is an important factor when hand-ling a huge amount of data and in ensuring

the acceptance of data warehouse application

users. However, it is not the only aspect to keep

in mind when deciding on a data warehouse

environment. The option of starting small and

increasing step-by-step as demands grow can

be a decisive factor as well as the high availa-

bility of the application.

P E R F O R M A N C E M E E T S S O L U T I O N S E X P E R T I S E

Oracle Database on Fujitsu PRIMERGY Cluster Delivers World Record Results

with SAP Business Intelligence Data Mart Standard Application Benchmark

Such demands urged Fujitsu and Oracle spe-

cialists to think about an advanced concept that

aligns performance with scalability and high

availability. The one and two-node benchmark

results already revealed the excellent scalability

of our PRIMERGY servers in combination with

the Oracle Database and Oracle Real Appli-

cation Clusters, says Joachim Witte, Senior

Director, Global Fujitsu SAP Competence Center.

Thus it was almost a logical next step to conti-

nue tests with more server nodes.

The following test series underlined the positive

results of the first sequence and has resulted

in a new world record. On the three-node

Fujitsu PRIMERGY RX 300 S5 cluster

throughput was increased to 900,309 query

navigation steps (3), while a further all-time high

of 1,165,742 query navigation steps (4) was

achieved by the four-node server cluster. So

Fujitsu and Oracle were once again happily ableto jointly announce a new world record on

December 1, says Joachim Witte.

The extremely successful benchmark series is

much more than a performance boost. In fact,

it is a perfect example of outstanding perfor-

mance that is obtained and embedded in using

an intelligent concept. In comparison to tradi-

tional clusters, the dynamic solution designed

by experts from the Global Fujitsu SAP Com-

petence Center in Walldorf (Germany) also

offers high system availability because Oracle

RACs monitor the environment and automati-cally migrate affected processes to other nodes

should a particular server fail. Furthermore,

customers have the option of starting small with

just one server node and growing step-by-step

as demands and data volumes increase.


23/82

Benchmark SAP on Oracle with Fuj itsu 21

The successful combination of cost-efficient

server technology and innovative database andcluster technology from Oracle also underlines

that Fujitsu PRIMERGY x86 servers today

are starting to match the high performance of

enterprise servers and are thus also suitable as

a migration path for enterprise warehousing

applications from end-of-life Unix-based

systems.

Note:

The latest Fujitsu results on the SAP BI-D

Standard Application Benchmark have been

certified under the following certificationnumbers:

2009036 (1) (single node), 2009037 (2)

(two nodes) and 2009045 (5) (four nodes).

Results as of October 14, 2009. For more

details, see http://www.sap.com/benchmark

(1) The SAP BI-D Standard Application

Benchmark performed on August 8, 2009

by Fujitsu in Walldorf, Germany has been

certified (certification number 2009036)

with the following data: Throughput/hour

(query navigation steps): 320,363.

Operating system, central server: SuSeLinux Enterprise Server 10. RDBMS:

Oracle Database 10g Real Application

Clusters (RAC). Technology platform

release: SAP NetWeaver 7.0 (non-

Unicode). Configuration: Central server:

Fujitsu PRIMERGY RX300-S5, 2 proces-

sors / 8 cores / 16 threads, Intel Xeon

Processor X5570, 2.93 GHz, 64 KB L1

cache and 256 KB L2 cache per core,

8 MB L3 cache per processor, 96 GB

main memory.


Benchmark performed on August 11, 2009

by Fujitsu in Walldorf, Germany has been

certified (certification number 2009037)

with the following data: Throughput/hour

(query navigation steps): 609,349.

Operating system all servers: SuSE Linux

Enterprise Server 10. RDBMS: Oracle Data-base 10g Real Application Clusters (RAC).

Technology platform release: SAP Net-

Weaver 7.0 (non-Unicode). Configuration:

2 servers (2 active nodes): Fujitsu PRI-

MERGY RX300-S5, 2 processors / 8 cores

/ 16 threads, Intel Xeon Processor X5570,

2.93 GHz, 64 KB L1 cache and 256 KB

L2 cache per core, 8 MB L3 cache per

processor, 96 GB main memory.


Benchmark performed on October 17,

2008 by IBM in Rochester, MN, USA wascertified on October 31, 2008 (certification

number 2008063) with the following data:

Throughput/hour: 182,112 query naviga-

tion steps. Operating system, Central server:

i 6.1. RDBMS: DB2 for i 6.1. Platform

release: SAP NetWeaver 7.0 (non- Unicode).

Configuration: Central server: IBM Power

System 570, 4 processors / 8 cores / 16

threads, POWER6, 5 GHz, 128 KB L1

cache and 4 MB L2 cache per core, 32 MB

L3 cache per processor, 128 GB main

memory.



2009 by Fujitsu in Walldorf, Germany

has been certified (certification number

2009044) with the following data:

Throughput/hour (query navigation steps):

900,309. Operating system, all servers:

SuSE Linux Enterprise Server 10. RDBMS:

Oracle Database 10g Real Application

Clusters (RAC). Technology platform re-

lease: SAP NetWeaver 7.0 (non-Unicode).

Configuration: 3 servers (3 active nodes):

Fujitsu PRIMERGY RX300-S5, 2 proces-

sors/8 cores / 16 threads, Intel Xeon

Processor X5570, 2.93 GHz, 64 KB L1

cache and 256 KB L2 cache per core,

8 MB L3 cache per processor, 96 GB main

memory.

So Fujitsu and Oracle

were once again happily

able to jointly announce

a new world record.

Joachim Witte,

Senior Director,Global FujitsuSAP Competence Center.


24/82

22



2009 by Fujitsu in Walldorf, Germany has

been certified (certification number

2009045) with the following data:

Through-put/hour (query navigation

steps): 1,165,742. Operating system all

servers: SuSE Linux Enterprise Server 10.

RDBMS: Oracle Dtabase 10g Real

Application Clusters (RAC). Technology

platform release: SAP NetWeaver 7.0

(non-Unicode). Configuration: 4 servers

(4 active nodes): Fujitsu PRIMERGY

RX300-S5, 2 processors / 8 cores / 16

threads, Intel Xeon Processor X5570, 2.93GHz, 64 KB L1 cache and 256 KB L2

cache per core, 8 MB L3 cache per pro-

cessor, 96 GB main memory.

About Fujitsu

Fujitsu is a leading provider of IT-basedbusiness solutions for the global marketplace.

With approximately 175,000 employees

supporting customers in 70 countries, Fujitsu

combines a worldwide corps of systems and

services experts with highly reliable computing

and communications products and advanced

microelectronics to deliver added value to

customers. Headquartered in Tokyo, Fujitsu

Limited (TSE:6702) reported consolidated

revenues of 4.6 trillion yen (US$47 billion)

for the fiscal year ended March 31, 2009.

For more information, please see:http://www.fujitsu.com

About Fujitsu Technology Solution

Fujitsu Technology Solutions is the leading

European IT infrastructure provider with a

presence in all key markets in Europe, the

Middle East and Africa, plus India, serving

large-, medium- and small-sized companies

as well as consumers. With its Dynamic Infra-

structures approach, the company offers a full

portfolio of IT products, solutions and services,

ranging from clients to datacenter solutions,Managed Infrastructure and Infrastructure-as-

a-Service. Fujitsu Technology Solutions employs

more than 10,000 people and is part of the

global Fujitsu Group. For more information,

please see: ts.fujitsu.com/aboutus

SAP, SAP NetWeaver and all SAP logos are

trademarks or registered trademarks of SAP

AG in Germany and in several other countries.

All other product and service names mentioned

are the trademarks of their respective companies.

Benchmark SAP on Oracle with Fuj itsuuj


25/82

23Benchmark SAP on Oracle Database 11 g Release 2 with Fuj itsu

O R A C L E D A T A B A S E 1 1 g D E L I V E R S W O R L D R E C O R DR E S U L T F O R A T W O - T I E R C O N F I G U R A T I O N O N S A PB U S I N E S S I N T E L L I G E N C E - D A T A M A R T S T A N D A R DA P P L I C A T I O N B E N C H M A R K

Oracle Press Release

Oracle Database 11gRunning on New Fujitsu

PRIMERGY RX600 S5 Server Raises the

Performance Bar for Business Intelligence

Application Workloads

Redwood Shores, Calif. March 30, 2010

News FactsOracle Database 11g set a world-record result

for a two-tier configuration on the SAP

Business Intelligence-Data Mart (BI-D)

standard application benchmark(1). On a

system comprised of one Fujitsu PRIMERGY

RX600 S5 server, equipped with the new

Intel Xeon X7560 2.26 GHZ processor,

Oracle Database 11g Release 2 on Linux deli-

vered a world-record result of 854,649 query

navigation steps per hour on the two-tier SAP

BI-D standard application benchmark.

Supporting QuotesWith this benchmark result, Oracle Database

11g Release 2 demonstrates performance leader-

ship on two-tier SAP BI-D standard application

benchmark and provides a solid proof-point

for customers seeking the best foundation for

their business intelligence applications, said

Juan Loaiza, senior vice president, Systems

Technology, Oracle.

Supporting Resources

Oracle Database

Oracle Real Application Clusters

Oracle Benchmarks About the SAP BI-Data Mart Standard

Application Benchmark

About Oracle

Oracle (NASDAQ: ORCL) is the worlds most

complete, open, and integrated business soft-

ware and hardware systems company. For more

information about Oracle, please visit our

Web site at http://www.oracle.com.

Trademarks

Oracle and Java are registered trademarks of

Oracle and/or its affiliates. Other names may

be trademarks of their respective owners. SAP,

SAP NetWeaver and all SAP logos are trade-

marks or registered trademarks of SAP AG in

Germany and in several other countries.Other names may be trademarks of their res-

pective owners.

Results as of March 30, 2010. For more infor-

mation, please visit

http://www.sap.com/benchmark.

The SAP BI Data Mart (BI-D) Standard

Application Benchmark performed on March

07, 2010 by Fujitsu in Walldorf, Germany,

has been certified with the following data:

(1) Throughput/hour (query navigationsteps): 854,649

CPU utilization of central server: 96%

(1 node active: 96%)

Operating system, central server: SuSE Linux

Enterprise Server 10

RDBMS: Oracle Database 11g

Technology platform release: SAP

NetWeaver 7.0 technology platform

(non-Unicode)

Configuration: 1 server (1 active node):

Fujitsu Primergy RX600-S5, 4 processors/32

cores/64 threads, Intel Xeon Processor

X7560, 2.26 GHz, 64 KB L1 cache and

256 KB L2 cache per core, 24 MB L3 cache

per processor, 128 GB main memory

Certification number: 2010011.

Contact Info:

Teri Whitaker

Oracle

[email protected]

Kristin Reeves

Blanc & Otus

[email protected]


26/82

24

The key results are discussed below

How does the combination of SAP and Oracle

running on Sun compare with the competition?

For comparison purposes we can use the

benchmark results for a 2-CPU Intel system

and an 8-CPU AMD system. For these we have

recent results for Windows with SQL Server,

Linux with MaxDB, and Solaris with Oracle,which can be used for comparison.

First, lets take a look at the 2-CPU results.

All results in the above diagram were measured

with servers that each had 2 Intel Xeon

x5570 CPUs with 8 cores and 16 threads.

The speed increase over SQL Server is 11%

when compared with Windows/SQL Server

and 20% for Linux/MaxDB.

The difference becomes more apparent in larger

systems. This demonstrates Suns and Oracles

considerable experience in the operation of

large systems. The available results for the 8-

processor AMD systems show a speed increase

for Oracle/Sun in the range of 21%-25%.

With 48 cores and 48 threads, these tested

systems have three times as many threads as

the 2-processor systems. The combination of

Oracle and the Solaris operating system is

the best combination for large SAP systems. It

offers the best load distribution over the many

threads.

If we analyze the current processor roadmaps,

we notice a trend towards more threads for all

manufacturers. Because future 2-processor

systems will also have more threads, customers

will benefit from this combination even with

smaller systems.

High-End Results

Since July 2008 the list of 2-tier SAP SD

benchmarks has been headed by a Sun M9000

system. Naturally, this benchmark used Oracle

Database 10g for the database as this ensures

optimum use of the large processing power of

the 64 CPUs with 256 cores and 512 threads.

The figure of 39,100 SAP SD users is based on

the SAP non-Unicode character set.

O R A C L E O N S O L A R I S :

B R E A K I N G S A P B E N C H M A R K R E C O R D SIn multiple SAP benchmark tests, Oracle Database on the Solaris operating system has

once again demonstrated its outstanding performance and scaling capabilities.

(Add each relevant database to diagram)


27/82

Benchmark SAP on Oracle-Solaris 25

New Benchmark Version

Since 1, January 2009, there was a newtwo-tier SAP ERP 6.0 Enhancement Package

4 (Unicode) Standard Sales and Distribution

(SD) Benchmark. This new release has higher

CPU requirements and fewer users compared

to the previous two-tier SAP ERP 6.0

(non-Unicode) Standard Sales and Distribution

(SD) Benchmark. For example: The new version

of the benchmark requires Unicode and the

Application layer uses a new encoding; the

response time was cut in half, form 2 seconds

to 1 second. Please refer to:

http://www12.sap.com/solutions/benchmark

/pdf/January2009newsfaq.pdf for more informa-tion.

So far Oracle has performed two benchmarks

on Enterprise Servers with the new SAP SD

benchmark based on SAP Enhancement

Package 4 for the SAP ERP 6.0 application

(Unicode). These two results jointly top the list

of best results. A Sun M9000 system with 32

CPUs (128 cores and 256 threads) was measured

first, achieving 17,430 SAP SD users. A good

month later Oracle outperformed this result.

A Sun M9000 system with twice the number

of CPUs achieved 32,000 SAP SD users.

This result highlights the SPARC Enterprise

M9000 servers performance leadership at the

high-end of SAP SD Standard Application

Benchmark and underscores the optimal per-

formance of SAP ERP running on SPARC64

systems with the Solaris OS regardless of the

encoding methodology used in the application.

SAP with Oracle RAC

Many SAP customers are looking for a way

of migrating expensive hardware to low-cost

standard servers. At the same time they expect

the same system availability. With Oracle RAC

it is possible to increase availability and use

low-cost Intel servers.

Running the Solaris 10 operating system on a

cluster of four Sun Blade X6270 server modules,each with two Intel Xeon x5570 quad-core

2.93 GHz processors (8 cores, 16 threads),

Oracle Database and Oracle Real Application

Clusters achieved 13,718 SAP SD-Parallel

Benchmark users, outperforming by 66 per-

cent the previous best eight processor two-tier

SAP SD Standard Application Benchmark

result with the SAP Enhancement Package 4

for SAP ERP (Unicode) held by Microsoft

SQL Server (2).

This result also delivered exceptional one-,

two- and four-node results on the Sun BladeX6270, demonstrating the outstanding scala-

bility of Oracle Database and Oracle Real

Application Clusters. When measured from

one-, two- and four-node configurations, as

documented and certified by SAP, Oracle

Database with Oracle Real Application Clusters

demonstrated near-linear scalability, achieving

3,800(3), 7,220(4) and 13,718(1) SAP SD-

Parallel Benchmark users, respectively, while

providing a high level of performance and

availability. No other vendor has delivered

results showing this level of performance

or scalability on the Intel Xeon x5570platform.

Additionally, the single-node result with

Oracle Real Application Clusters delivers more

performance-per-processor and per-core on the

SAP SD Standard Application Benchmark

running SAP Enhancement Package 4 for SAP

ERP 6.0 (Unicode) and outperforms the best

SAP SD two-tier results from IBM DB2(5)

and Microsoft SQL Server(6) with this bench-

mark result, utilizing the same number of cores.

Oracle Database

continues to outpace

the competition in

performance and

scalability. This bench

mark demonstrates

once again that Real

Application Clusters

scales performance fo

real-world complex

business applications.

Together, Oracle and

Sun are a winning

combination.

Juan Loaiza,

Senior Vice President,Systems Technology,Oracle.


28/82

26

AP Virtualization? Yes with Solaris Containers

On September 10, SAP Standard Application Bench-mark certification was also achieved with a virtualized

system landscape using Oracle Database 10g and a

Sun Fire X4270 including the virtualization solution

Oracle Solaris Container (7). The SAP Sales and

Distribution (SD) Standard Application Benchmark,

achieved on the Solaris 10 operating system with

Oracle Solaris Container (using 8 vCPUs), was sup-

ported by the out-standing performance and scaling

capabilities of the Oracle and Sun combination run-

ning SAP.

In comparison with another benchmark (8) on the

similar hardware with the same CPU type (2 proces-sors/8 cores/16 threads, Intel Xeon Processor x5570,

2.93 GHz) from Fujitsu, which included the virtua-

lization solution VMware and database MaxDB 7.8,

the Sun Container virtualization solution was 36

percent better than VMware. The other major advan-

tages of the Solaris Containers virtualization technology

are, firstly, the price Solaris Containers is included

in Solaris like the advanced new file system ZFS and,

secondly, the superior support model. Oracle Solaris

Containers have been supported by SAP productive

systems for over four years and complete support, from

hardware to database, is now available from a single

source Hardware. Software. Complete!

http://fixunix.com/

Results as of February 1, 2010. Source: SAP,

www.sap.com/benchmark.

(0) The SAP Sales and Distribution (SD) StandardApplication Benchmark: Sun M9000, 32 processors / 128cores / 256 threads, SPARC64 VII, 2.88 GHz, 17,430 SAPSD Benchmark users. 95,480 SAPS. SAP EnhancementPackage 4 for SAP ERP 6.0 (Unicode), Oracle Database 10g,Solaris 10; SPARC Enterprise Server SAP certificationnumber: 2009038.

(01) The two-tier SAP Sales and Distribution (SD)Standard Application Benchmark: IBM System 550, 4processors / 8 cores / 16 threads, Power6, 5GHz, 3,752SAP SD Benchmark users; ; 20,520 SAPS SAP Enhance-ment Package 4 for SAP ERP 6.0 (Unicode), DB2 9.5;AIX 6.1; SAP certification number: 2009023.

(02) The SAP Standard Application SD (Sales and Distri-

bution) Benchmark is a two-tier ERP business test that isindicative of full business workloads of complete orderprocessing and invoice processing, and demonstrates theability to run both the application and database software ona single system. The SAP Standard Application SD Bench-mark represents the critical tasks performed in real-worldERP business environments.

(1) The SAP Sales and Distribution-Parallel (SDParallel)Standard Application Benchmark: 4 active nodes: SunBlade X6270, 2 processors / 8 cores / 16 threads, IntelXeon Processor X5570, 2.93 GHz, 13,718 SAP SD-ParallelBenchmark users. 75,770 SAPS, SAP Enhancement Package4 for SAP ERP 6.0 (Unicode), Oracle Database 10g RealApplication Clusters (RAC), Solaris 10; SAP certificationnumber: 2009041.

(2) The two-tier SAP Sales and Distribution (SD)Standard Application Benchmark: HP ProLiant DL785G6, 8 processors / 48 cores / 48 threads, Six-Core AMDOpteron 8439 SE, 2.8 GHz,8,280 SAP SD Benchmark,45,350 SAPS, SAP Enhancement Package 4 for SAP ERP6.0 (Unicode), SQL Server 2008; Windows Server 2008Enterprise Edition; Certification number: 2009035.

(3) The SAP Sales and Distribution-Parallel (SDParallel)Standard Application Benchmark: 1 active nodes: SunBlade X6270, 2 processors / 8 cores / 16 threads, IntelXeon Processor X5570, 2.93 GHz, 3,800 SAP SDParallel,20,750 SAPS, SAP Enhancement Package 4 for SAP ERP6.0 (Unicode), Oracle Database 10g Real Application

Clusters (RAC), Solaris 10; SAP certification number:2009039.

(4) The SAP Sales and Distribution-Parallel (SDParallel)Standard Application Benchmark: 2 active nodes: SunBlade X6270, 2 processors / 8 cores / 16 threads, IntelXeon Processor X5570, 2.93 GHz, 7,220 SAP SDParallel,39,420 SAPS, SAP Enhancement Package 4 for SAP ERP6.0 (Unicode), Oracle Database 10g Real ApplicationClusters (RAC), Solaris 10; SAP certification number:2009040.


29/82

27

(5) The two-tier SAP Sales and Distribution (SD)Standard Application Benchmark: IBM System 550, 4processors / 8 cores / 16 threads, Power6, 5GHz, 3,752 SAPSD Benchmark, 20,520 SAPS, SAP Enhancement Package4 for SAP ERP 6.0 (Unicode), DB2 9.5; AIX 6.1; IBMSystem 550, 4 processors/8 cores/16 threads, POWER6,5.0GH Certification number: 2009023.

(6) The two-tier SAP Sales and Distribution (SD)Standard Application Benchmark: ProLiant BL460c G6,2 processors / 8 cores / 16 threads, Intel Xeon Processorx5570, 2.93 GHz, 3,415 SAP SD Benchmark, 18,670SAPS, SAP Enhancement Package 4 for SAP ERP 6.0(Unicode), SQL Server 2008; Windows Server 2008Enterprise Edition; IBM System 550, 4 processors/8cores/16 threads, POWER6, 5.0GHz Certification number:2009031.

(7) The SAP two-tier Sales and Distribution (SD)Standard Application Benchmark: Sun Fire X4270,2 processors / 8 cores / 16 threads, Intel Xeon ProcessorX5570, 2.93 GHz, 2,800 SAP SDParallel, 15,320 SAPS,SAP Enhancement Package 4 for SAP ERP 6.0 (Unicode),Oracle Database 10g, Solaris 10, Segmentation/CPUutilization in non-global zones: 1 non-global zone using 8virtual CPUs SAP certification number: 2009034.

(8) The SAP two-tier Sales and Distribution (SD)Standard Application Benchmark: Fujitsu PRIMERGYModel RX300 S5, 2 processors/8 cores/16 threads, IntelXeon Processor x5570, 2.93 GHz 2,056 SAP SDParallel,11,230 SAPS, SAP Enhancement Package 4 for SAP ERP

6.0 (Unicode), MaxDB 7.8, SuSE Linux Enterprise Server10 on VMware ESX Server4.0, Segmentation/CPU utiliza-tion in virtual machines: 1 virtual machine (VM) using 8virtual CPUs Segmentation/CPU utilization in non-globalzones: 1 non-global zone using 8 vir

Volume19 En

Documents

Transcript of Volume19 En