Joseph W. O‘Leary DataVard, Inc. Shrink your database with automated housekeeping
Dec 05, 2014
© 2014 DataVard # 1
Joseph W. O‘Leary
DataVard, Inc.
Shrink your database with automated housekeeping
© 2014 DataVard # 2
§ Reduce system size in regards to Logs and Temporary Data substantially
§ Improve loading performance § Be better prepared for future
challenges like migration to SAP HANA® platform
§ Transparency § Reduce data cleansing effort
§ Minimum effort, maximum benefit § ERNA functionalities (i.e. cockpit
for more transparency) § Experience of DataVard
consultants § Price
§ Result of Analysis (DataVard BW Fitness Test): 1.308 GB temporary data and logs blocked nearly half of the system size 25 % database size reduction
§ Deletion of 750 GB unnecessary data (PSA, ChangeLog, Cube compression, RFC Logs)
§ Reduction of workload: only § 1 workday per month for
housekeeping
Customer Example
Customer Profile An international leader in specialist publishing in the area Science, Technology, Medicine.
“With the help of ERNA we were able to substantially reduce the effort for data cleansing, with minimum effort and realizing savings of 25 % of the system’s size. We are now well prepared for future challenges like SAP HANA®.”
Key figures Objectives Highlights
Database Size Reduction: 25% through Housekeeping
© 2014 DataVard # 3
Clean up your system Size reduction example (Housekeeping and NLS)
183 183
998 321
918
780.3
650
325
312
156
0
48.1
0
500
1000
1500
2000
2500
3000
3500
Heute mit OutBoard und ERNA
OutBoard
Cube data
ODS data
Other data
Temporary data
Master data
Before After
-68%
-15%
-50%
-50%
Total DB space saved of 43%!
© 2014 DataVard # 4
Use
r ha
ppin
ess
TCO
& d
ata
acce
ss
TCO
Smart Data Management
§ Performance optimization, Tuning
§ In-memory § Ensure SLAs are met
GOALS TACTICS
§ Use appropriate storage: Archiving, NLS, Smart data access
§ Set up central policies
§ Define policies § Set up housekeeping § Automation
Information “at your fingertips”
speed and high availability is key.
Keep & store, but reduce costs.
Purge, delete, housekeeping
Hot Data Business critical data Data required for reporting and planning
Cold Data / Old Data Aged data, history Infrequent, rare use Need to keep (legal, internal, industry requirements)
Dead Data Technical data (e.g. logs, protocols, PSA) Redundant data
© 2014 DataVard # 5
5%
15%
15%
9%
11%
32%
5% 5% 3%
Master data
Temporary data
Other data
PSA data
Changelog data
ODS data
Cube E data
Cube F data
Cube D data
Step 1: Fitness Test Typical distribution of data in a BW system
Comments: § Data you report on is
only 13-17% of the system size
§ Temporary data is subject to housekeeping (BALDAT, RS*DONE, ...)
§ Use the HANA sizing
report as a 1st indication (OSS note 1736976).
§ Create a plan for the
data lifecycle (data load to data exit)
“Only 12% of all data in BW is actually used” Source: Forrester research
© 2014 DataVard # 6
Housekeeping activities
n Application log n Batch log n IDoc tables (EDI40, EDIDS) n qRFC, tRFC n Job-Tables (TBTCO, TBTCP etc.) n Change & Transportsystem n Spool data (TST03) n Table Change Protocols n Batch Input Folders n Alert Management Data (SALRT*) n Old short dumps n Batch input data
Netweaver
Scope of Housekeeping
n Unused customers n Unused vendors n Phantom change documents n Phantom texts
ERP
n PSAs & Change Logs n Request logs & tables (RSMON*
and RS*DONE) n Unused dimension entries n Unused master data n Cube & Aggregate compression n Temporary database objects n NRIV buffering n Table buffering n BI-Statistics n Process Chain Log n Errorlogs n Unused Queries n Empty partitions n BI Background processes n Bookmarks n Web templates
Business Warehouse
© 2014 DataVard # 7
7 reasons housekeeping is not done
Organizational Functional
n Unclear responsibilities Basis or Application-Team?
n Low visibility What happens when its not done is not transparent
n No dedicated resources Will management pay for cleanup?
n Tools spread out in the system Many small programs that can be found in many SAP notes
n Not enough functionality If there is a program it often only allows keep or delete, no archiving
n Too high risk Deleting permanently is risky
n No transparency no tool that says if cleanup is running correct and in full
© 2014 DataVard # 8
Central cockpit for all activities
© 2014 DataVard # 9
Recycle Bin makes more aggressive deletion possible
Ease discussions about retention times.
Instead of deleting data, you can temporarily store it in a compressed recycle bin, from which it can be deleted at a later date.
© 2014 DataVard # 10
Recycle Bin compression (example PSA)
Same retention time using less space
Today
6 m
onth
s P
SA
14 d
ays
PS
A
15 days- 6months compressed in recycle bin (Quick Restore possible)
Benefit:
ERNA Recycle Bin
Automated deletion
© 2014 DataVard # 11
Less work through one central access
Keep your whole SAP Landscape in good condition
ERNA runs either stand-alone in a single SAP System or is installed on a central instance to automatically housekeep the full system landscape
© 2014 DataVard # 12
Mass processing leads to lower implementation effort
By grouping objects, that are to be house kept, the implementation effort is kept low.
New objects can be automatically included to existing groups, by using wildcards and dynamic patterns.
© 2014 DataVard # 13
Calendar & Scheduler
Ensure, that the planned housekeeping activities are really being implemented.
There are multiple monitoring tools: the calendar view shows all planned tasks within the full SAP Landscape, where ERNA is running.
© 2014 DataVard # 14
Clean up your system in 3 steps
Start Run Deploy 1 32
§ Prepare Project
§ Kick off workshop
§ Analyze potential
§ Scope definition based on BW Fitness Test and archiving roadmap
§ Target storage definition
§ OutBoard™ (NLS / ERNA) installation
§ Initial customizing
§ Housekeeping setup
§ Object adjustments
§ Deletion and retrieval tests
§ Performance tests
§ Setup of ongoing archiving / deletion
§ Knowledge Transfer
§ Initial Housekeeping
§ Result validation
§ Optional: Tablespace reorganization
§ Project Sign-off & Support
§ Ongoing Housekeeping done by customer
© 2014 DataVard # 15
Your presenter:
Joseph W. O‘Leary Product Manager ILM DataVard, Inc.
Request a demo here: www.erna.datavard.com
© 2014 DataVard # 16
Who is DataVard
§ Specialized in Data Management for SAP
§ Customers range from SMEs (60 users) to Fortune 500 (e.g. Allianz, BASF, KPMG, Roche, Nestle)
§ Focus on Data Management and ABAP development
§ SAP and ABAP only § SAP certified solutions for BW Nearline storage
and housekeeping § Partnership with SAP in consulting (e.g. SLO) § Partnership with SAP in development (e.g. ILM)
Success
Experience
Focus
© 2014 DataVard # 17
Copyright DataVard Inc. 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 DataVard GmbH. The information contained herein may be changed without prior notice. DataVard and OutBoard are trademarks or registered trademarks of DataVard GmbH and its affiliated companies. SAP, R/3, SAP NetWeaver, SAP BusinessObjects, SAP MaxDB, SAP HANA and other SAP products and services mentioned herein as well as their respective logos are trademarks or registered trademarks of SAP AG in Germany and other countries. 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. These materials are provided by DataVard GmbH and its affiliated companies (“DataVard") for informational purposes only, without representation or warranty of any kind, and DataVard shall not be liable for errors or omissions with respect to the materials. The only warranties for DataVard 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.
Copyright