SAP Solution Manager 7.

1 Sizing Guide

SAP Solution Manager Sizing Guide

Document Version 2.2 September 2013 Valid for SAP Solution Manager 7.1 SP10

SAP AG Dietmar-Hopp-Allee 16 69190 Walldorf Germany T +49/18 05/34 34 24 F +49/18 05/34 34 20 www.sap.com

Copyright 2007 SAP AG. All rights reserved. No part of this publication may be reproduced or transmitted in any form or for any purpose without the express permission of SAP AG. The information contained herein may be changed without prior notice. Some software products marketed by SAP AG and its distributors contain proprietary software components of other software vendors. Microsoft, Windows, Outlook, and PowerPoint are registered trademarks of Microsoft Corporation. IBM, DB2, DB2 Universal Database, OS/2, Parallel Sysplex, MVS/ESA, AIX, S/390, AS/400, OS/390, OS/400, iSeries, pSeries, xSeries, zSeries, z/OS, AFP, Intelligent Miner, WebSphere, Netfinity, Tivoli, and Informix are trademarks or registered trademarks of IBM Corporation in the United States and/or other countries. Oracle is a registered trademark of Oracle Corporation. UNIX, X/Open, OSF/1, and Motif are registered trademarks of the Open Group. Citrix, 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. Java is a registered trademark of Sun Microsystems, Inc. JavaScript is a registered trademark of Sun Microsystems, Inc., used under license for technology invented and implemented by Netscape. MaxDB is a trademark of MySQL AB, Sweden. Disclaimer Some components of this product are based on Java. Any code change in these components may cause unpredictable and severe malfunctions and is therefore expressively prohibited, as is any decompilation of these components. Any Java Source Code delivered with this product is only to be used by SAPs Support Services and may not be modified or altered in any way. These materials are subject to change without notice. These materials are provided by SAP AG and its affiliated companies ("SAP Group") for informational purposes only, without representation or warranty of any kind, and SAP Group shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP Group products and services are those that are set forth in the express warranty statements accompanying such products and services, if any. Nothing herein should be construed as constituting an additional warranty. SAP, R/3, mySAP, mySAP.com, xApps, xApp, SAP NetWeaver, 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 in several other countries all over the world. 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.

Typographic Conventions
Type Style Example Text Represents Words or characters quoted from the screen. These include field names, screen titles, pushbuttons labels, menu names, menu paths, and menu options. Cross-references to other documentation. Emphasized words or phrases in body text, graphic titles, and table titles. Technical names of system objects. These include report names, program names, transaction codes, table names, and key concepts of a programming language when they are surrounded by body text, for example, SELECT and INCLUDE. Output on the screen. This includes file and directory names and their paths, messages, names of variables and parameters, source text, and names of installation, upgrade and database tools. Exact user entry. These are words or characters that you enter in the system exactly as they appear in the documentation. Variable user entry. Angle brackets indicate that you replace these words and characters with appropriate entries to make entries in the system. Keys on the keyboard, for example, F2 or ENTER.

Icons
Icon Meaning Caution Example Note Recommendation Syntax

Example text

EXAMPLE TEXT

Example text

Example text

<Example text>

EXAMPLE TEXT

September 2013

Contents
1 2 3 4 SUMMARY ........................................................................................................................19 SAP SOLUTION MANAGER ARCHITECTURE ...............................................................19 FACTORS THAT INFLUENCE PERFORMANCE ............................................................21 SIZING METHODOLOGY .................................................................................................25 4.1 Approach .................................................................................................................25 4.2 Managed Systems characteristics ........................................................................25 4.2.1 Size of the managed landscape ...................................................................25 4.2.2 Solution type of the managed landscape ......................................................26 4.2.3 Assumptions..................................................................................................27 4.2.4 SAP Solution Manager Scenario ..................................................................28 4.2.5 Number of Business Processes ....................................................................28 4.2.6 Business Processes Monitoring objects .......................................................28 4.2.7 Number of Concurrent active users ..............................................................29 4.2.8 Number of EEM Robots ................................................................................29 4.2.9 Number of Service Desk documents ............................................................29 4.2.10 TREX.............................................................................................................30 4.3 Solution Manager Sizing KPIs ...............................................................................31 4.4 SAP Solution Manager's sizing .............................................................................33 4.4.1 Introscope SmartStor ....................................................................................33 4.4.2 Solution Manager Database .........................................................................33 4.4.3 Database Monthly growth .............................................................................37 4.4.4 Cluster Consideration ...................................................................................38 4.4.5 CPU and Memory .........................................................................................39 4.4.6 Network .........................................................................................................43 4.4.7 Impact on Managed systems ........................................................................43 PERFORMANCE SETTINGS ............................................................................................44 5.1 Solution Manager Settings ....................................................................................44 5.2 Wily IS EM setup and tuning .................................................................................45

Sizing Guide

September 2013

Summary

This document describes how to estimate hardware resources that are necessary to implement different scenario on SAP Solution Manager7.1 System. It provides rules, guidelines, and best practices for the definition of hardware requirements like main memory, disk space and CPU. It will also identify the required cluster configuration for Solution Manager and for Introscope infrastructure to implement SAP Solution Manager. This document should be used with the SAP Solution Manager Sizing Toolkit spreadsheet which implements the formulas described in this document.

SAP Solution Manager Architecture

The SAP Solution Manager runs in a separate central system, to which all other SAP systems are connected. Systems that are administered using the SAP Solution Manager are referred to as "managed systems". The sizing of SAP Solution Manager does not include the sizing of the managed systems, but it is strongly dependent on the size of the managed systems, and on the number of managed systems connected to the Solution Manager. The architecture of SAP Solution Manager relies on the following elements Agents technology SAP Solution Manager Introscope Enterprise Manager (EM)

Agents technology 3 agents are installed on the managed systems: SAP Host Agent (SAP HA) SAP DAs (SAP DA) running at the Host level of the managed systems. CA Introscope Bytecode Injection Agent (CA IS BCI Agent) running in the SAP J2EE instances of the managed systems.

Those agents send data on regular pre-defined time interval from the managed systems to Solution Manager.

Sizing Guide

19

September 2013

Solution Manager SAP Solution Manager runs on top a SAP NetWeaver dual stack. It collects and aggregates data sent by the SAP DAs. In a simple environment, this NW 7.0 Web Application Server double stack manages all data collection and aggregation. In a more complex environment, a NetWeaver cluster must be implemented to manage large landscape of managed systems.

Introscope Enterprise Manager Introscope EM is a server running on a central place that collects, performs calculations on, and stores metrics reported by multiple agents. In a simple Introscope environment, one single standalone Enterprise Manager collects, persists, and processes all the agent metrics, then supplies the resultant data for viewing in the Introscope Workstation or WebView browser instances. In a more complex environment, SAP recommends a cluster scenario based on MOM and Collectors: Enterprise Managers in the role of Collectors can be clustered so that their collected metrics data is compiled in a single Manager of Managers (MOM) EM, which provides a unified view of all the metrics to the connected Workstation and WebView instances. Each Collector will be associated to a Smart Data Store. In this configuration, all EMs in a MOM / Collector setup should be located in the same LAN.

This document will consider: The sizing for the ABAP stack requirements, the J2EE stack requirements and the Introscope EM requirements. The impact of the Agents on the managed systems.

This document will not consider the performance tuning of SAP Solution Manager.

House Keeping Considerations Housekeeping in SAP Solution Manager is a process to clean and aggregate the data collected from the managed systems. This document makes the assumption the defaulthousekeeping settings are taken and not changed. Changing the default-housekeeping will invalidate the sizing recommendations contained in this document.

Sizing Guide

20

September 2013

Factors that influence performance

Type of Scenario implemented in SAP Solution Manager Number and role of concurrent Users accessing Solution Manager Solution Landscape including hardware configuration of the managed Landscapes and Software infrastructure of the managed landscapes Product information Load of the Solution

The sizing of SAP Solution Manager depends on the following factors:

SAP Solution Manager Scenario

The factors influencing the performance depend on the different scenarios that will be implemented on Solution Manager. The following scenarios are considered in this document: Basic Configuration (including Root Cause Analysis and EWA) Root Cause Analysis provides standardized tools to enable a systematic and targeted end-to-end component localization of the root cause of IT problems. As Root Cause Analysis is a mandatory scenario in order to implement SAP Solution Manager, it also includes the Basic Configuration of Solution Manager. Collaboration The Collaboration section of the SAP Solution Manager is the point of access to SAP support services, including remote services, on-site services, self-services, and Best Practice documents. Service Desk The Service Desk of the SAP Solution Manager offers a complete infrastructure for organizing and operating solution-wide support. Change Request Management The objective of Change Management is to carry out changes economically and in a timely manner with minimal risk. Change Request Management includes processes for managing change requests, project management, and change logistics. Implementation and Upgrade (includingtesting) The Implementation and Upgrade scenarios support the implementation of customer projects and the definition of template projects and upgrade projects. It includes roadmaps, editor for creating and maintaining business blueprints, access to the Implementation Guides (IMG), testing scenarios, monitoring and distributing Customizing. Solution Monitoring Technical Monitoring The technical monitoring scenario monitors systems performance, availability, configuration and exception. It includes also specific product monitoring such as BI, PI and connection Monitoring. It covers also jobs monitoring use-cases. Business Process Monitoring This scenario deals with monitoring of ore business processes including Business Process Analytics.

Sizing Guide

21

September 2013

End User Experience Monitoring The EEM scenario controls the availability and the performance of the solution from user point of view.

Data Volume Management The Data Volume Management scenario defines how to manage data growth by avoiding data creation, deletion and archiving data.

Concurrent active Users

Concurrent users are working simultaneously or concurrently in the system. They consume system resources at the same time. Active users are defined as running a given number of business processes in a given time period. We distinguish between three categories of active users, because these categories represent typical activity patterns of users - A Low User or occasional user processes on average ten interaction steps an hour or one every six minutes. - A Medium User processes in average 120 interaction steps an hour or one every 30 seconds. Most users will have a comparable profile - A High User processes an average of 360 interaction steps an hour or one every 10 seconds. This is typical for power users working in call centers or doing data entry.

In addition to those three categories, during the sizing process of Solution Manager, we will distinguish later in this document between different types of User depending on their role on Solution Manager. For example, a Service Desk User will not consume the same level of resources than an Administrator using only Solution Monitoring and Root Cause Analysis.

Solution Landscape
The following elements of the solution landscape of the managed systems connected to SAP Solution Manager will influence the sizing Number of Systems. Number of hosts (or virtual hosts) per system. A SAP DA is installed on each host (physical or virtual). Starting with SAP Solution Manager 7.1 SP5, DA support OTF (On The Fly) mode. This does not impact the sizing of SAP Solution Manager as DA nodes and DA instances have the same impact on SAP Solution Manager. Number of server nodes or instances per host.

The combination of systems/hosts/instances will determine the number of agents reporting data to Solution Manager.

Sizing Guide

22

September 2013

Product information
The volume of data reported by the managed systems is highly dependent on the software components running on the managed systems. The type of products (Portal, XI, ECC) running in the solution, will indicate the nature of those components and thus will indicate the volume of data sent by the agents running on the managed systems.

Load of the Solution

The products of the solution gives already a good indication on the data sent by the agents, but to get the correct amount of data, a user sizing or a throughput sizing on the managed systems is needed. To simplify the process, we assume that the size of the managed system goes into one of the following category: M L XL XXL

The load of the solution will influence the number of data collected in SAP Solution Manager.

Standard Sizing for SAP Solution Manager

The standard sizing recommendations for End-to-End Root Cause Analysis including Basic Configuration in SAP Solution Manager are the following:

Data Base

Disk Space installation

SAPs

Memory

SAP Solution Manager Introscope EM

80 GB (DB) 40 GB (SmartStor)

20 GB 250 MB

2000 2000

8 GB 4 GB

Sizing Guide

23

September 2013

Remarks on standard settings

This sizing should always be checked against your landscape configuration with the guidelines described in the rest of the document. The following rules in this document are used to validate that the standard configuration will be enough to support the operation of E2E RCA in your landscape. We recommend installing Solution Manager and Introscope EM two different servers for a productive installation. In case of a non-productive installation, Solution Manager and Introscope EM can be installed on the same server. In general, those sizing recommendations correspond to the implementation of the Basic configuration (EWA, RCA, Collaboration) scenario for a maximum of 5 large Solutions (including different roles: Dev, QA, Prod) for different product types.

Sizing Guide

24

September 2013

Sizing methodology

4.1 Approach
The sizing methodology for Basic configuration in SAP Solution Manager follows the steps described below: Characteristics of the managed systems landscape a. Size of the managed landscape. b. Product information of the managed Landscape. c. System Load of the managed Landscape. d. Solution Manager's scenario. e. Number of concurrent users. Computation of the sizing KPIs corresponding to the characteristics of the managed systems. Calculation of SAP Solution Manager's resources. a. Disk Size. b. Cluster configuration. c. Hardware Sizing (CPU and Memory). d. Impact on managed systems.

4.2 Managed Systems characteristics

4.2.1

Size of the managed landscape

The first questions to answer when considering the solution manager environment are: How many systems do I have to connect to my Solution Manager? How many server nodes and hosts are supporting those systems?

Parameters Number of systems #Systems

Description For each system, the minimal DB size for SAP Solution Manager and Introscope EM is computed later in this document.

Number of hosts per system #Servers

One SAP DA (node or instance) will be installed per host (virtual or physical) The number of SAP DAs will be used later in the sizing process. One CA IS BCI Agent will be installed per Java Server node. The number of CA IS BCI Agents will be used later in the sizing process.

Number of Java Server nodes per systems #Java Server nodes

Sizing Guide

25

September 2013

4.2.2

Solution type of the managed landscape

The next question to answer is about the amount of data that will be collected on solution Manager. It depends on the products running in the solution and the expected load on the solution. The load can to be either measured on existing systems or estimated with a T-shirt Sizing project. Please refer to SAP Quick-Sizer reference: http://service.sap.com/QuickSizer for more information.

Parameters Type of the solution

Description The amount of data collected by the RCA Solution Operation tool depends on the type of the solution. A portal solution will generate iViews metrics that are not reported in an XI solution. On the contrary, an XI solution will produce, queue information metrics.

Expected Load per system

The load on the system depends on the number of users and the transactional throughput on the system. This measure depends in general on the role assigned to the system. For example, a 'production' system will most likely produce L or XL load of data. In the same way, a sand-box, or QA system will be of type S or M.

The expected load should be classified in one of the following category. Load per system M L XL XXL Description / Guidelines Dev system Production system with medium volume of users or transactions Production system for large volume of users or transactions Production system for extra large volume of users or transactions

Sizing Guide

26

September 2013

4.2.3

Assumptions

In order to simplify the process, we assume that the number of data reported per usage type could be classified into two categories: High Low

So for each product supported by RCA in SAP Solution Manager, the following approximation is performed regarding the number of metrics reported to Introscope EM (#metrics) and the number of Extractor Framework records collected by Solution Manager (#Ext Fmk Records).

Usage type WebAS6.40/7.00 (ABAP) WebAS6.40/7.00 (J2EE) ECC5.0 / 6.0 Portal incl. KMC TRex7.0x/7.1X BI 3.1/3.5/7.0x/7.1x XI 3.x/7.0x/7.1x CRM4.0 CRM5.0 / 5.2 /2007 SRM 6.0 Duet 1.0 MDM5.5 SP6 P1 / 7.1 xMII 11.5 SAP Business byDesign Mobile Infrastructure WEB Dispatcher 7.10 SAP Solution Manager 7.0x SAP HANA SAP Data Service 4.0 Standalone DB Database

#metrics Low High High High Low Low High Low High Low Low Low Low High Low Low Low Low Low Low Low

#Ext Fmk Records High Low High Low Low High High High High Low Low Low Low High Low Low High High High High Low

Sizing Guide

27

September 2013

4.2.4

SAP Solution Manager Scenario

The next KPI to find is the list of Solution Manager's scenarios that will be implemented. By default, the Root Cause Analysis/Basic configuration has to be implemented. The rest of the scenarios are optional as they depend on your Solution Manager's requirements.

4.2.5

Number of Business Processes

For the Solution Documentation scenario, you need to find out how many business processes will be involved in those scenarios.

4.2.6

Business Processes Monitoring objects

For the Business Process monitoring scenario, the following metrics will influence the sizing of SAP Solution Manager: Monitoring objects Frequency of Monitoring objects collection Number of Solution monitored Number of end users

You will find below the default values of those KPIS.

Description Number of Monitoring objects with 5 min frequency Number of Monitoring objects with 10 min frequency Number of Monitoring objects with 15 min frequency Number of Monitoring objects with 30 min frequency Number of Monitoring objects with 45 min frequency Number of Monitoring objects with 60 min frequency Number of Monitoring objects with 90 min frequency Number of Monitoring objects with 120 min frequency Number of Monitoring objects with 12 hrs frequency Number of Monitoring objects with 1 day frequency Number of BPO workcenter/Alert inbox concurrent end users Number of BP Analytics end users (concurrently working) Number of solutions

KPI #monitoring_objects1 #monitoring_objects2 #monitoring_objects3 #monitoring_objects4 #monitoring_objects5 #monitoring_objects6 #monitoring_objects7 #monitoring_objects8 #monitoring_objects9 #monitoring_objects10 #endusers2 #endusers1 #nbsolutions

Default 300 0 0 0 0 0 0 0 30 40 10 10 1

Sizing Guide

28

September 2013

4.2.7

Number of Concurrent active users

The number of concurrent active users per Solution Manager's scenarios depends on the type of scenario implemented on your Solution Manager, For this reason, we consider two Roles of Users when performing a sizing of SAP Solution Manager. Role 1: Users running the following scenarios Service Desk, CharM, Collaboration.

Role 2: Users running the following scenario Implementation Upgrade

Root Cause Analysis and Technical Monitoring are not taken into account due to the fact that they are not user's centric application. The main factors influencing the performance for those scenarios are the load of data produced by the managed systems.

4.2.8

Number of EEM Robots

In End User Experience Monitoring scenario, we need to consider the number of robots that will be implemented.

4.2.9

Number of Service Desk documents

In Service Desk scenario, the KPI to consider is the number of tickets created per day.

Sizing Guide

29

September 2013

4.2.10

TREX

In TREX/ scenario, the KPI to consider are Business Object Size, Replication and Search and Server Topology Business Object Size This sizing guide assumes a certain size for business objects. For all these objects we assume an average instance size of 4 kB in the database. This figure is an average based on measurements of several business object types of different applications. The basis for this assumption is that the data of the business object is maintained in two languages. For more languages this figure must be increased accordingly. This sizing guideline assumes a compression ratio of 1:2 from business object size in the database to index size in main memory.

File Size This sizing guide assumes a certain size for files which are kept in the file system of the Trex server. The average size of a file in the file system is assumed to be 400 kB. A file contains text in one language. This sizing guideline assumes a compression ratio of 1:20 from file size on the file system to index size in main memory.

Object Size Business Object File

Size on Disk (*) 4 kB 400 kB

Size in Main Memory 2 kB 20 kB

Replication and Search This sizing guide assumes that the two main tasks, replication and search (see section 1.1), are running separately at different time slots. The search response time may increase if replications are running simultaneously.

Server Topology Because of the various complex aspects of a general sizing, the forecast of the system CPU resources is done for an average CPU utilization of 65%. The power of CPUs for SAP applications is measured in SAPS. An overview of the different CPUs and their corresponding SAPS values is provided in the SAP benchmarking overview: For each search provider and each business object 1 connector is required to perform indexing and search tasks.

Sizing Guide

30

September 2013

4.3 Solution Manager Sizing KPIs

Based on the characteristics of the managed systems landscape, the two tables below are used as input to estimate the KPIs of the Solution Manager's sizing. This table below is used to identify the relevant sizing KPIs for one System instance of a specific product, based on the Load (observed or estimated) and the type of product.

System Load

M Low High

L Low High

XL Low High

XXL Low High

Sizing Units Number of Introscope metrics per CA IS BCI Agents #metrics per CA IS BCI Agents Sizing Units Number of Introscope metrics per SAP DAs #metrics per SAP DAs Sizing Units Number of Extractor Framework records per day per Managed Systems #Ext Fmk Records

CA IS BCI Agents
500 1500 750 4000 1000 7500 4000 30000

SAP DAs
250 1000 500 1500 750 2000 3000 8000

Managed Systems
5000 25000 10000 50000 15000 10000 60000 400000

By repeating this step for all systems of the solution landscape attached to your Solution Manager, you will find the total number of data collected and aggregated in SAP Solution Manager.

Sizing Guide

31

September 2013

The table below is used to classify the concurrent active users depending on their roles and categories. As described earlier in this document, two roles are relevant for the Sizing's project. For each role, we consider three families of Users: Low, Medium and High as described in the table below:

User #Role 1 Low Activity User #Role 1 Medium Activity User #Role 1 High Activity User #Role 2 Low Activity User #Role 2 Medium Activity User #Role 2 High Activity User

Description 10 dialog steps per hour or less for Service Desk, CharM or Collaboration scenario. 2 dialog steps per minute or less for Service Desk, CharM or Collaboration scenario. 6 dialog steps or more per minute or less for Service Desk, CharM or Collaboration scenario. 10 dialog steps per hour or less for Implementation and Upgrade scenario. 2 dialog steps per minute or less for Implementation and Upgrade scenario. 6 dialog steps or more per minute or less for Implementation and Upgrade scenario.

Sizing Guide

32

September 2013

4.4 SAP Solution Manager's sizing

4.4.1
Root Cause Analysis The size of the Introscope SmartStor is given by adding the size for all systems as described in the formulas below. This size is a fixed size as the data are aggregated by the IS EM. Introscope SmartStorSize (MB) (#Java Server nodes * #metrics per CA IS BCI Agents + #Servers * #metrics per SAP DAs)* 0,09

Introscope SmartStor

4.4.2

Solution Manager Database

The size of the Solution Manager Database can be identified by using the formulas corresponding to the scenarios implemented for your Solution Manager. Basic + Root Cause Analysis BI Database Size (MB) INITIAL + (#Systems * #Ext Fmk Records) * 0.12

End-User Experience Monitoring (Technical Operations) BI Database Size (MB) CharM + Implementation DB SIZE (MB) #Role 1 Low Activity User / 300 + #Role 1 Medium Activity User / 30 + #Role 1 High Activity User / 10 ) * 3600 * Number of work hour per day * Workdays per year * 5,12 / 1000 / 1000 #Role 2 Low Activity User / 300 + #Role 2 Medium Activity User / 30 + #Role 2 High Activity User / 10 ) * Workdays per year * 531738 / 10000 / 1000 Service Desk DB SIZE (MB) Solution documentation DB SIZE (MB) #BPNumber*200 #number of tickets per day * Workdays per year *5.12 / 5 / 1000 (#nb of Robots *1000) * 0.12

Data Volume Management DB SIZE (MB) #Systems*50*12

Sizing Guide

33

September 2013

Business Process Monitoring DB SIZE (MB) 60/5*(#monitoring_objects1)*0,03*24 + 60/10*(#monitoring_objects2)*0,03*24 + 60/15*(#monitoring_objects3)*0,03*24 + 60/30*(#monitoring_objects4)*0,03*24 + 60/45*(#monitoring_objects5)*0,03*24 + 60/60*(#monitoring_objects6)*0,03*24 + 60/90*(#monitoring_objects7)*0,03*24 + 60/120*(#monitoring_objects8)*0,03*24 + (#monitoring_objects9)*0,03 + (#monitoring_objects10)*0,03

Technical Monitoring Input table #product #number of metrics 200000 300000 300000 300000 300000 300000 #percenta ge of BI records 60 % 70 % 70 % 70 % 70 % 70 % #alerts #host dependent metrics 60 % 60% 60 % 60 % 60 % 60 % #metric type 20 20 20 20 20 20

All Products (default) SAP HANA SAP Netweaver CE 7.1 WebAS 6.40/7.0x/7.1x (ABAP) WebAS 6.40/7.0x/7.1x (J2EE) PI 3.x/7.0x/7.1x

30 30 30 30 30 30

Granularity #granu #granu1 #granu2 #granu3 #granu4 #granu5 #granu6 #granu7 Granu Type Minute Hour Day Weekly weekly with hour profile Monthly Monthly with hourly profile 1/7 1/7 * 24 1/30 (1 / 30) * 24 Explanation 1440 minutes/day 24 hours per day Value 1440 24 1 0.143 3.429 0.033 0.8

Sizing Guide

34

September 2013

Lifetime #lifetime #lifetime 1 #lifetime2 #lifetime3 #lifetime4 #lifetime5 #lifetime6 #lifetime7 LifeTime Type Minute Hour Day Weekly weekly with hourly profile Monthly Monthly with hourly profile Explanation 8 days history 1 month history 100 days One year history 1 month history 4 years history 1 month history Value 8 31 100 52 31 47 31

KPIs #number_of_metric_types = 800 + #product_metric_type * #product #number_BI_metrics = #systems * #number_of_metrics * #percentage_of_BI_records* #host_dependant_metrics + * #number_of_vhosts

#systems * #number_of_metrics * #percentage_of_BI_records * (1-#host_dependant_metrics)

Alert Alert tables size (KB) = #system * #alerts * 20 Aggregates Aggregates table size (KB) = #number_of_metric_types * 220 DSO Data growth per day (bytes) = #granu1 * (#number_BI_metrics) + #granu2 * (#number_BI_metrics) + .+ #granu7 * (#number_BI_metrics) Assumption: The DSO size is included in the BI table size.

Sizing Guide

35

September 2013

BI #number_of_BI_record = #granu1 * #lifetime1 *(#number_BI_metrics)+ #granu2 * #lifetime1 *(#number_BI_metrics)+ .+ #granu7 * #lifetime7 * (#number_BI_metrics)

Data Volume (Bytes stored in BI) = #number_of_BI_records * 160 Byte + #number_BI_metrics *(8(dimtable)+ 76(SID table)) + #number_of_metric_types * (8 + 132 ) + #number_of_metric_types * (12 + 32 + 132) ( Metric name) + #number_of_systems*(44 + 72 + 52 + 72 + 52 + 52 + 72 + 52+ + #number_of_vhosts * (16 + 132 + 132 + 132) + #number_of_vhosts * (12 + 132 + 132) (Object) + #number_of_java_nodes * (16 + 24 + 132 + 132)

Assumptions: - The "object" dimension is equal to the "host" dimension. - The metric name dimension is equal to the metric type dimension.

DB SIZE (GB)

BI Data Volume + Aggregates + Alert tables size

TREX Input Disk Sizing: NumberBOs = Total number of business objects to be loaded NumberFiles = Total number of files to be loaded Constants: ConstDiskSizeBO [kByte] = 4 ConstDiskSizeFile [kByte] = 400

DB SIZE (GB)

2 * 10-6 * ( NumberBOs * ConstDiskSizeBO + NumberFiles * ConstDiskSizeFile )

Sizing Guide

36

September 2013

4.4.3
Basic

Database Monthly growth

The BI Aggregation mechanism will add an additional monthly growth of 1% of the total size of the BI Database. Please refer to the BI archiving documentation to manage this growth.

Technical Monitoring No monthly data base growth in this scenario is expected.

All other scenarios: CharM, Implementation and Service Desk The data base growth for those scenarios corresponds to the same values computed by formulas mentioned above.

Sizing Guide

37

September 2013

4.4.4

Cluster Consideration

The following formulas have to be taken into consideration when clustering Solution Manager and Introscope EM to support E2E RCA for complex and large Solution landscape. You should consider clustering when the number of SAP DAs is greater than 400 or/and the number of CA IS BCI Agents is greater than 50.

Root Cause Analysis Rules One Server Node can handle 500 SAP DAs. Description Add one additional Server node every 500 connected SAP DAs.

A IS Collector can handle 500 000 metrics.

Add one additional IS Collector every 500 000 metrics. For example, for a XL/High system it represents 70IntroscopeAgents. Add one dialog work process for every system producing 100000 Extractor Framework Records per day. Add one dialog work process for every group of systems producing 100000 Extractor Framework Records per day. Depending on the number of dialog work processes derived from this formula, you may consider to add additional ABAP instances to your Solution Manager.

1 dialog work process can handle 100000 Extractor Framework Records per day. In addition, we recommend adding 1 dialog work process every 30 users.

The number of Dialog and batch Work process is given by the Sizing Toolkit

Minimum: 20 + 1 dialog work process / 30 users + 1 dialog work process / 1000000 1 batch work process = 1 dialog work process / 3

Sizing Guide

38

September 2013

4.4.5

CPU and Memory

The CPU and memory configuration can be identified by using the formulas corresponding to the scenarios implemented for your Solution Manager. Basic configuration Each cluster element added in Solution Manager to support the managed landscape has to follow the rules below: Memory (GB) Solution Manager main instance Additional Solution Manager server node Introscope EM Additional IS Collector 8 4 4 4 CPU (SAPS) 2000 2000 2000 2000

#SAPS_per_system 0,03

#SAPS_per_server 0,04

#mem_per_system 4

#mem_per_server 5

Size M L XL XXL

#factor 0,5 1 2 5

Memory (GB) Application Server LF * (#system * #factor * #mem_per_system + #server * #factor * #mem_per_server) * 3/5 LF * (#system * #factor * #mem_per_system + #server * #factor * #mem_per_server) * 2/5

CPU (SAPS) LF * (#system * #factor * #saps_per_system + #server * #factor * #saps_per_server) * 3/5 LF * (#system * #factor * #saps_per_system + #server * #factor * #saps_per_server) * 2/5

Database Server

Note LF: Load Factor = 65%

Sizing Guide

39

September 2013

Business Process Monitoring Memory (GB) IF(((300*#endusers2)+(150*#endusers1 ))*(1-(0,05*#nbsolution))> 10000; 10000;((300* endusers#2)+(150* #endusers1))*(1-(0,05*#nbsolution))) CPU (SAPS) 60/5*(#monitoring_objects1)*0,0070170327975599*24 + 60/10*(#monitoring_objects2)*0,0070170327975599*24 + 60/15*(#monitoring_objects3)*0,0070170327975599*24 + 60/30*(#monitoring_objects4)*0,0070170327975599*24 + 60/45*(#monitoring_objects5)*0,0070170327975599*24 + 60/60*(#monitoring_objects6)*0,0070170327975599*24 + 60/90*(#monitoring_objects7)*0,0070170327975599*24 + 60/120*(#monitoring_objects8)*0,0070170327975599*24 + (#monitoring_objects9)*0,0070170327975599 + (#monitoring_objects10)*0,0070170327975599

Role 1 User (CharM) Memory (GB) Application Server (#Role 1 Low Activity User +#Role 1 Medium Activity User + #Role 1 High Activity User ) * 62 / 1000 (#Role 1 Low Activity User + #Role 1 Medium Activity User + #Role 1 High Activity User ) * 6.2 / 1000 CPU (SAPS) (#Role 1 Low Activity User / 300+ #Role 1 Medium Activity User /30+ #Role 1 High Activity User / 10 ) * 730 * 4/5 (#Role 1 Low Activity User / 300+ #Role 1 Medium Activity User / 30+ #Role 1 High Activity User / 10 ) * 730 * 1/5

Database Server

Role 2 User (Implementation) Memory (GB) Application Server (#Role 2 Low Activity User +#Role 2Medium Activity User+#Role 2High Activity User )*18,75/1000 (#Role 2 Low Activity User +#Role 2Medium Activity User+#Role 2High Activity User )*2,55/1000 CPU (SAPS) (#Role 2 Low Activity User / 300+ #Role 2 Medium Activity User / 30+ #Role 2 High Activity User / 10 ) * (200000/100/20)*3)4/5 (#Role 2 Low Activity User /300+ #Role 2 Medium Activity User /30+ #Role 2 High Activity User / 10 ) * (20000/100/20)*3)1/5

Database Server

Note : For Service Desk and CharM scenarios users are concurrent Users.

Solution Documentation Memory (GB) Database Server (#BPNumber *0,6)/ 500 CPU (SAPS) (#BPNumber * 2.5)

Sizing Guide

40

September 2013

Service Desk Memory (GB) Application Server Database Server #number_of_users * 62 / 1000 #number_of_users * 6,2 / 1000 CPU (SAPS) (#number_of_tickets_per_day *10 /24 / 3600) * 730 (#number_of_tickets_per_day *10 / 24 /3600) *730 * 1/5

Assumptions: The number of dialog steps to complete a ticket is: 10.

End User Experience Monitoring Memory and CPU depends in number of robots which are considered as Virtual Hosts. One Java Server node can manage 500 Virtual Host.

Technical Monitoring #SAPS_per_system 0,02 #SAPS_per_server 0,03

#mem_per_system 4

#mem_per_server 5

Size M L XL XXL

#factor 0,75 1 2 5

Memory (GB) Application Server #system * #factor * #mem_per_system + #server * #factor * mem_per_server * 3/5 #system * #factor * #mem_per_system + #server * #factor * mem_per_server * 2/5

CPU (SAPS) (#system * #factor * #saps_per_system + #server * #factor * #saps_per_server) * 3/5 (#system * #factor * #saps_per_system + #server * #factor * #saps_per_server)*2/5

Database Server

Shared Mem (KB) = #number_of_metric_types * 18

Sizing Guide

41

September 2013

TREX CPU (SAPS) Input CPU Sizing: ConcurrentUsers = Average number of concurrent users during peak times QueriesPerUser = Average number of queries per user in one hour ConnectorsPerQuery = Average number of connectors/query Constants: ConstCpuPerConnector 65 [SAPS/ConnectorCall/sec] = 25.3 ConstCpuAvgUtil [%] =

Required CPU Power [SAPS] ConnectorCallsPerSec * ConstCpuPerConnector / (ConstCpuAvgUtil/100) = ConcurrentUsers * QueriesPerUser * ConnectorsPerQuery * 38.92/3600

MEMORY Input Memory Sizing: NumberBOs = Total number of business objects to be loaded NumberFiles = Total number of files to be loaded Constants: ConstMemSizeBO [kByte] = 2 ConstMemSizeFile [kByte] = 20

Required Memory [GigaByte] 2 * 10-6 * ( NumberBOs * ConstMemSizeBO + NumberFiles * ConstMemSizeFile )

Sizing Guide

42

September 2013

4.4.6

Network

For best usage, network responses times should be adapted to the different needs of communication used between the different elements of the RCA architecture in Solution Manager. Network Collector to MOM Requirement When a MOM requests data from a Collector, the roundtrip response must be less than 500 ms. The network transfer rate between Solution Manager and the Managed systems must be greater or equal than 100Mbit/s.

DAs agents (CA IS BCI Agents, SAP DAs) to Solution Manager systems

4.4.7

Impact on Managed systems

SAP DAs, CA IS BCI Agents and Extractor tasks have a very limited impact on the managed systems The table below summarizes the resources required by RCA in SAP Solution Manager's activities (CA IS BCI Agents, SAP DAs, Extractor jobs) on the managed systems: The numbers below are given for a XL/High system.

Impact on managed systems Disk (MB) Memory (MB) Network CPU SAP DA 7.1X: 500 MB per Host (or Virtual Host) where an SAP DA is installed. 256 MB in AIX environment or 128 MB for other environments per Host (or Virtual Host) where an SAP DA is installed. 3% usage for a 100Mbit/s Network bandwidth. 5 % CPU average increase per Host (or Virtual Host) belonging to the Managed Systems.

Sizing Guide

43

September 2013

Performance Settings

5.1 Solution Manager Settings

Number of Java dispatcher threads. Java dispatcher threads are used for the p4 communication between the SAP DAs and Solution Manager. The number of Java dispatcher threads depends on the number of SAP DAs connected to the Solution Manager Instance. This number should be set according to the following formula: number_of_Dispatcher_Thread (ThreadManager->MaxThreadCount) = number_of_SAP DA_agents * 1.75

Default value of

MaxThreadCount

After an installation of Solution Manager, the default value of the parameter MaxThreadCount in the Java Dispatcher configuration is set to 70. This setting allows by default to connect maximum 40 agents.

Maximum Heap Size for Dispatcher The java heap size of the Dispatcher should be set to the following value in order to avoid timeout connection issues:

Java Max Heap Size (-Xmx) = Number of SAP DAs * 1.28

Number of RFC listener RFC listeners are used for the JCO/RFC communication between the ABAP and Java stacks of Solution Manager. The number of RFC listeners is given by the following formula: number_of_RFC_listener = number_of_Background_process + 1 / number_of_Java_Server_Node

Sizing Guide

44

September 2013

5.2 Wily IS EM setup and tuning

Collector to MoM clock drift limit. MoM and Collector clocks need to be synchronized to within AT MOST three seconds. JVM setttings The JVM Heap Size (-Xmx) must be large enough to handle all active metrics. The newSize parameter must be adapted accordingly and the Xms should be equal to the Xmx parameter. You will find below a typical java command line parameters used to start a Collector: Xms4096m -Xmx4096m -Xmn800m -verbose:gc -Dswing.volatileImageBufferEnabled=false Djava.awt.headless=true Disk Requirement EM Storage File system used for Enterprise Manager files storage baselines.db, and traces.db should be a local disk and not a network file system (NFS). Using NFS for EM storage could result in performance degradation that can be covered to some extent by having enough RAM available as FS cache, or by adding another EM. Cluster Consideration In a Wily IS MoM/Collector environment, every Collector must have a dedicated disk I/O subsystem for SmartStor with no other processes competing for it. This needs to be set appropriately in the IntroscopeEntrepriseManager.propertiesfile for every Collector and EM like described in the example below:

# The directory that the EM will use for data storage introscope.enterprisemanager.smartstor.directory=data introscope.enterprisemanager.smartstor.directory.archive=data/archive introscope.enterprisemanager.smartstor.dedicatedcontroller=true

Only the Collector requires a dedicated SmartStor database and dedicated controller. MOM machines also have a SmartStor instance, but due to the vastly smaller metrics load, are able to house the SmartStor instance on the same disk as other MOM components. Providing a separate disk for each SmartStor AND setting the dedicated controller property to true affects the total number of metrics an Enterprise Manager can handle because these allow for better sharing of disk resources. To determine if a machine being considered for use as SmartStor is a single dedicated disk or drive, you may need to determine if the machine has multiple controllers (same as multiple hard drives). It's important to understand that multiple partitions on the same drive share a controller, which is not an appropriate environment for the SmartStor instance. You can use commands like du (for disk usage) on Unix/Linux or Windows Device Manager to determine whether two drives are logically different or physically different. It's critical that the drives are physically different.

Sizing Guide

45