Public
ANP203 – Size and Fine-tune the
SAP BusinessObjects Business Intelligence
platform
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 2 Public
Speakers
Bangalore, October 5 - 7
Narashimman K S
Las Vegas, Sept 19 - 23
Harjeet Judge
Barcelona, Nov 8 - 10
Henry Banks
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 3 Public
Disclaimer
The information in this presentation is confidential and proprietary to SAP and may not be disclosed without the permission of
SAP. Except for your obligation to protect confidential information, this presentation is not subject to your license agreement or
any other service or subscription agreement with SAP. SAP has no obligation to pursue any course of business outlined in this
presentation or any related document, or to develop or release any functionality mentioned therein.
This presentation, or any related document and SAP's strategy and possible future developments, products and or platforms
directions and functionality are all subject to change and may be changed by SAP at any time for any reason without notice.
The information in this presentation is not a commitment, promise or legal obligation to deliver any material, code or functionality.
This presentation is provided without a warranty of any kind, either express or implied, including but not limited to, the implied
warranties of merchantability, fitness for a particular purpose, or non-infringement. This presentation is for informational
purposes and may not be incorporated into a contract. SAP assumes no responsibility for errors or omissions in this
presentation, except if such damages were caused by SAP’s intentional or gross negligence.
All forward-looking statements are subject to various risks and uncertainties that could cause actual results to differ materially
from expectations. Readers are cautioned not to place undue reliance on these forward-looking statements, which speak only
as of their dates, and they should not be relied upon in making purchasing decisions.
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 4 Public
Agenda
Overview
BI Platform conceptual tiers
Platform
Infrastructure & Architecture & Sizing
Hardware: Physical & Virtual
Web Tier
Splitting, Sizing & distributing Web-components
Intelligence Tier
Central Management Server & System Database(s)
Disk Tuning & File Repository Servers
Processing Tier
Configuring the processing servers
Fine-tuning the details
Clients Tools & Data Access
Usage & Efficiency of reporting
Semantic Layer & Federation
Supportability & Monitoring Tools
BI4.2 performance enhancements
Public
Overview
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 6 Public
BI Platform Conceptual Tiers
Public
Platform considerations
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 8 Public
Infrastructure requirements Architecting BI systems
Sizing & Landscape
Demands of BI applications means bursty & dynamic allocations that hit processor & memory.
Do you have enough CPU power?
Do you have enough RAM Memory? Be generous for Java applications.
– Consult the Sizing Guides and professional services. Revisit your Sizing frequently!
Are your server processes logically distributed across nodes?
– Be prepare to scale-out your systems, & cluster across new servers.
Design your system include load-balancing & fail-over tolerances:
– Have extra services on each node & across other hosts
Benefits of fewer, larger machines (good for processor & memory management) is offset by local IO
requirements & availability.
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 9 Public
Hardware: Physical or Virtual?
Know about best-practices for virtualized environments
Ensure strict CPU reservations and Memory pre-allocation:
– Avoid use of shares, limits, affinity, or other artificial mechanisms to
manage capacity across hosts.
Ensure virtualization hosts can handle the aggregate requirements
from multiple tenants:
– Hardware needs to be properly designed to cope with the IO, in
particular Disk activity & Network.
o Consider SSD & RAID for best disk speed, & multiple Network Interface
Cards.
VMs offers advantages for HA & DR among things, but traceability
can be an issue:
– Watch out for communications timeouts due to overloaded VM
network bridge (software, as opposed to hardware).
Public
Web Tier
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 11 Public
Splitting, Sizing & distributing the web components Failover, Load-balance & performance
Accelerate static content:
Add an Apache web server to the deployment
Split the Web & Application Tiers to improve the user-experience.
Add Multiple Tomcat application servers, approximately every 500 CAL.
Introduce software load-balancers into the topology.
Increase -Xmx via the Tomcat console (Default 2048MB to 10GB+ per node)
Increase java -Xmx parameter value in web.xml properties on the AppServer.
Increase Tomcat maxThreads to 500 from default 200.
Other settings:
Upgrade Tomcat & JVM versions (Tomcat 8 in BI4.2)
Evaluate proprietary JAVA & Tomcat .xml parameters (such as DEFLATE)
Maximize caching opportunities at browser, proxy, & web-server levels.
Public
Intelligence Tier
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 13 Public
System Database(s) Central Management Server, Auditing & Monitoring
BI is a bandwidth-heavy application
Beware: Latency from poorly provisioned CMS databases with have a
cascading effect!
A dedicated CMS DB running on separate hardware is recommended.
Increase the number of requested connections from default 14 to 50 threads.
Increase the CMS ‘Max number of objects in Cache’ in-line with actuals
– Cluster CMS when planning over 500 Active Concurrent users.
Make sure the CMS DB is available to provide low latency & high throughput
Check vendor-specific optimisations with your DBA, statistics are up to date.
– Monitor CMS response times, & check for long-running infostore queries.
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 14 Public
System Database(s) Central Management Server, Auditing & Monitoring
Configure Auditing setting to limit grown of Audit
database
– Reduce “Delete Events older that (days) parameter
– Only Audit events that you require
– Monitor Audit Thread Utilization (%)
o Consistent high % usage could indicate problem with
auditing system
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 15 Public
Disk Tuning the File Repository Servers
An underperforming FRS can starve the entire BI system
• I/O Bottlenecks on the disks & network cause severe effects.
– Your file-system should be tuned for ‘Write’ & ‘Read-after-write’
operations.
– Plan for capacity increases due to hot-backup & scheduling output.
Ensure good performance of inter-node communication links,
and use quality SAN storage.
For FRS, run checks on NAS connectivity (vendor-specific).
Use local, fast storage for WebI Cache & Temp-file directories.
– If response times are too low (<=10 ms) your environment will
slow.
Public
Processing Tiers
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 17 Public
Configuring the processing servers
Each server can happily consume 100GB of RAM
Consult the Sizing Companion Guide! Create and distribute additional processing & job servers across
nodes:
– WebI Processing, Visualization, Data Federation, Scheduling & Promotion, Monitoring & Auditing etc.
– Remember that Add-Ons for Lumira, Design Studio, Analysis Office etc. require significant additional sizing!
Split the Adaptive Processing Servers (APS) using the Configuration Wizard templates
– Adjust the application settings (i.e. capturing of details, events, frequency) for Monitoring & Platform
Search.
Refine configuration using APS Best Practices, to make further adjustments to the heap size (-Xmx cmd line)
– Do this for all Java servers (DSL_Bridge etc. up to 30GB each) & CR/Dashboard child sub-processes.
– Evaluate alternative Garbage Collection parameters in APS command line properties (parallelGC, newGC).
– WebI: Increase (or deselect) Upper & Maximum memory thresholds in Properties. Increase Binary Max
Stream & Character stream values etc.
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 18 Public
Fine-tuning the details
Co-locate the Processing Servers with the data-source in the same network
Schedule reports to cached instances, rather than refresh on-demand.
– Use Server groups to distribute schedules.
– Use Monitoring probes to watch system metrics & alert administrators in real-time.
– Adjust Idle document & connection timeouts, as required.
– Increase Heap values in ConnectionServer.cfg & MDAS.properties.
– Adjust MaxLovSize parameters in client descriptor.xml files.
– Consider use of LOVChunk & ResultChunkSIZE settings for BICS.
Public
Clients Tools & Data Access
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 20 Public
Clients Tools & Semantic Layer
Careful usage & efficiency of reporting
Complex query operators can have larger memory footprints: Be careful
when using Unions, Cartesian products, Full-scans, Grouping Sets etc.
Good query design includes:
Use query-stripping, drill, and guided navigation. Look to refactor multi-query
documents.
Use pre-defined and dynamic filtering, with default selection-options for variable
and parameters inputs.
Avoid 100s of heavy client-side calculations (e.g. Nested Sections with
Conditions).
Avoid very wide tables (50-200 objects) & data-dumping queries (500K-1M rows).
Activate Caches.
Evaluate viewer & editor preferences: HTML vs. Java. (Adjust memory heap
settings using JRE console.)
For big performance gains, keep your BW & HANA updated to coincide with
enhancement in BI releases.
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 21 Public
Data Access, Semantic Layer & Federation
Consider efficiency & proximity of reporting Database &
connected network speed
Ensure good relational Universe design & revisit:
– Index awareness (on keys rather than labels)
– Aggregate awareness (on summary tables)
– Delegated aggregation (measure projection)
– Use of Shortcut joins & views
– Tune Multisource universes & optimise the Federation join-engine
Edit connection properties to:
– Adjust Thread pool, Fetch & Array size, check advanced query parameters
(JOIN_BY_SQL for UNX, MAX_CONCURRENT_MEMORY_CONSUMING for MSU)
– Define safety-belts & timeouts to avoid data-dumping
– Configure hierarchy depth, members selection, exclude unassigned nodes.
– Avoid too many details or navigational attributes
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 22 Public
Supportability & Monitoring Tools
Tools
• BIPlatform CMC Monitoring & OSCol (SMDAgent)
• Platform support tool 2.0 & E2E traces with client plugin • http://service.sap.com/sap/support/notes/1861180
• Wily Introscope & Solution Manager
• ST0x transactions & BW BIPTools
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 23 Public
Supportability & Monitoring Tools
Utilities
• GLFViewer
• HTTPWatch
• JMeter, JVMMon
• WebIAdminTool
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 24 Public
BI4.2 SP03 - innovations for performance
2. User Group to Server Group affinity
4. Web Intelligence HTML parity with Java
5. Web Intelligence publication delivery rules
6. Suppress web app deployment 3. Pre caching of installers
1. Web Intelligence Parallel Queries
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 25 Public
SAP TechEd Online
Continue your SAP TechEd
education after the event!
Access replays of
Keynotes
Demo Jam
SAP TechEd live interviews
Select lecture sessions
Hands-on sessions
…
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 26 Public
Further information
Related SAP TechEd sessions:
ANP100 SAP BusinessObjects Lumira and SAP BusinessObjects Design Studio
ANP107 SAP BusinessObjects BI Platform 4.2 - Learn Pattern Books for Upgrading
ANP110 Building Stunning Analytics for SAP BusinessObjects Mobile
SAP Public Web
www.sapbi.com
www.sap.com/bisizing
scn.sap.com/community/bi-platform
SAP Education and Certification Opportunities
www.sap.com/education
Watch SAP TechEd Online
www.sapteched.com/online
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 27 Public
Please complete your
session evaluation for
ANP203.
Contact information:
Harjeet Judge, Narashimman K S, Henry Banks
SAP Analytics Product Management.
Feedback
© 2016 SAP SE or an SAP affiliate company. All rights reserved. 28 Public
© 2016 SAP SE or an SAP affiliate company. 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 SE or an SAP affiliate company.
SAP and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP SE (or an SAP affiliate
company) in Germany and other countries. Please see http://www.sap.com/corporate-en/about/legal/copyright/index.html for additional trademark information and notices.
Some software products marketed by SAP SE and its distributors contain proprietary software components of other software vendors.
National product specifications may vary.
These materials are provided by SAP SE or an SAP affiliate company for informational purposes only, without representation or warranty of any kind, and SAP SE or its
affiliated companies shall not be liable for errors or omissions with respect to the materials. The only warranties for SAP SE or SAP affiliate company 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.
In particular, SAP SE or its affiliated companies have no obligation to pursue any course of business outlined in this document or any related presentation, or to develop
or release any functionality mentioned therein. This document, or any related presentation, and SAP SE’s or its affiliated companies’ strategy and possible future
developments, products, and/or platform directions and functionality are all subject to change and may be changed by SAP SE or its affiliated companies at any time
for any reason without notice. The information in this document is not a commitment, promise, or legal obligation to deliver any material, code, or functionality. All forward-
looking statements are subject to various risks and uncertainties that could cause actual results to differ materially from expectations. Readers are cautioned not to place
undue reliance on these forward-looking statements, which speak only as of their dates, and they should not be relied upon in making purchasing decisions.