In-Memory Oracle BI Applications (UKOUG Analytics Event, July 2013)

T : +44 (0) 8446 697 995 E : enquiries@rittmanmead.com W: www.rittmanmead.com
In-Memory Analytics with Oracle BI Apps and Oracle Exalytics
UKOUG Analytics Event, London, July 2013
Mark Rittman, Technical Director, Rittman Mead
1Friday, 5 July 13

About the Speaker
•Mark Rittman, Co-Founder of Rittman Mead
•Oracle ACE Director, specialising in Oracle BI&DW
•14 Years Experience with Oracle Technology
•Regular columnist for Oracle Magazine
•Author of two Oracle Press Oracle BI books
•Oracle Business Intelligence Developers Guide
•Oracle Exalytics Revealed
•Writer for Rittman Mead Blog :
http://www.rittmanmead.com/blog
•Email : mark.rittman@rittmanmead.com
•Twitter : @markrittman
2Friday, 5 July 13

About Rittman Mead
•Oracle BI and DW platinum partner
•World leading specialist partner for technical excellence, solutions delivery and innovation in Oracle BI
•Approximately 50 consultants worldwide
•All expert in Oracle BI and DW
•Offices in US (Atlanta), Europe, Australia and India
•Skills in broad range of supporting Oracle tools:
‣ OBIEE
‣ OBIA
‣ ODIEE
‣ Essbase, Oracle OLAP
‣ GoldenGate
‣ Exadata
‣ Endeca
3Friday, 5 July 13

Oracle Business Intelligence 11g
•Oracle’s business intelligence platform, 11.1.1.7 release came out in April 2013
•Fourth major release of OBIEE 11g, with many new features + updated look and feel
•Enterprise BI platform centered around the Common Enterprise Semantic Model (RPD)
•Mobile BI apps, MS Office integration, ad-hoc,
dashboard and published reporting
•Built around Oracle Fusion Middleware
•Deployable on Windows, Unix, Linux
•Accessing a range of enterprise data sources
‣ Oracle and other RDBMSs
‣ Essbase and other OLAP servers
‣ Files, XML, web services
‣ ADF and SOA sources
‣ TimesTen in-memory database
4Friday, 5 July 13

Oracle BI Applications
•Packaged version of OBIEE that includes a data warehouse, and ETL mappings,
from E-Business Suite, Siebel, SAP and Peoplesoft
•Covers areas such as Financial Analytics, HR Analytics, Sales Analytics etc
•Built on the same technology as OBIEE 11g, plus ETL and administration tools
5Friday, 5 July 13

Pre-Defined ETL Routines from Oracle EBS, Peoplesoft, Siebel, JDE, SAP
•Integrated, conformed dimensional data warehouse
•Deployable on Oracle, MS SQL, IBM DB/2 and Teradata
•Uses Informatica PowerCenter for ETL, or now ODI11g
•Staging tables and presentation tables
•Allows modular deployment
•Lowest grain of information
•Prebuilt aggregates
•History tracking
•Indexing
6Friday, 5 July 13

BI Apps Data Warehouse Limitations
•Designed for “lowest common denominator” DB features
‣ No “out of the box” partitioning, MVs, compression optimization for PQ
•Based on traditional disk-based RBDMS technology
•Can often lead to slow reports, dashboards, limiting user acceptance
•Common issue - what can we do about it?
7Friday, 5 July 13

Oracle Exalytics : First In-Memory Engineered System for Analytics
• Relational, Multi-Dimensional and Unstructured data analysis available as a single engineered system
• Combination of in-memory hardware and optimized software versions
• Supports the Exadata and Big Data Appliance data management systems
Exalytics
In-Memory
Machine
Spans Relational, Multi-Dimensional, and Unstructured analysis,
combined with Financial & Operational Planning
‣ In-Memory Optimized Hardware
‣ In-Memory Oracle BI, TimesTen, Essbase, and Endeca
‣ Many In-Memory Software Innovations
Tightly-Integrated with Exadata, and
Big Data Appliance
8Friday, 5 July 13

Oracle Exalytics Benefits vs. Standard Hardware + Software
•Single supported stack of hardware + software : patching synchronized and tested across all components
•OBIEE, Essbase, TimesTen etc optimizations that are only available when deployed on Exalytics hardware
•Optimal selection of CPUs, RAM (DRAM), network connectors for a BI application tier
•Automatic in-memory caching of commonly-used aggregates - no manual tuning and selection
•Future platform for all Oracle BI products - EPM Suite, BI, Endeca, BI Apps
9Friday, 5 July 13

Exalytics as the Exa-Machine for OBIEE
•Runs the BI layer on a high-performance, multi-core, 1TB server
•In-memory cache used to accelerate the BI part of the stack
•If Exadata addresses 80% of the query performance, Exalytics addresses
the remaining 20%
‣ Consistent response times for queries
‣ In-memory caching of aggregates
‣ 40 cores for high concurrency
‣ Re-engineered BI and OLAP software
that assumes 40 cores and 1TB RAM
ERP/Apps DW
Oracle BI
In-Memory DB/Cache
10Friday, 5 July 13

Exalytics Under the Covers - How Does it Work?
•Exalytics brings together different technologies, which are still standalone products in their own right
•To harmonise and optimise their use within Exalytics, it utilises the following techniques:
‣ In-Memory Adaptive Data Mart - Using Oracle TimesTen for Exalytics, an in-memory RDBMS
‣ In-Memory Intelligent Result Cache
‣ In-Memory Cubes
•Some of these are genuine "secret sauce"
•New functionality and algorithms
•You can only get them through licensing Exalytics
•Others are descriptions of DW/BI strategies, or existing product functionality, extended to take advantage of
the capacity for processing in memory that Exalytics has
11Friday, 5 July 13

In-Memory Adaptive Data Mart
•Commonly-used aggregates are copied into Oracle TimesTen for Exalytics
•Past query patterns are analyzed and suitable aggregates recommended
•Oracle BI Server then uses these aggregates to make queries run faster
•Aggregates change over time in response to
changes in query patterns
•Tools are provided for managing and populating these aggregates
TimesTen BI Server
Exalytics
Aggregates
Data Warehouse
Detail-level
Data
12Friday, 5 July 13

Oracle TimesTen for Exalytics
•New version of TimesTen specifically for Exalytics (and only available with Exalytics)
•Support for analytic functions
‣ Perform all the processing at source
‣ Combine with being in-memory = should be very fast
•Column compression
‣ Whitepaper cites 5x
‣ Given the hardware capacity, we could seriously contemplate loading the whole Data Warehouse into
memory
‣ Opens up lots of interesting design potential
•We can load aggregates into TimesTen, leave base data at source, and use OBIEE’s Vertical Federation
capability to seamlessly report across both
‣ All hidden from the end-user, all they will know is that their reports run fast!
13Friday, 5 July 13

TimesTen and OBIEE Architecture
•Oracle BI Server communicates with TimesTen through TimesTen Client
•Summary Advisor, and nqcmd use Oracle BI Server to access TimesTen
•Typical single TimesTen database per Exalyics machine
‣ Max TimesTen database size around 300MB
- Due to need to set aside equal
Temp size for the Perm size selected
•Clustered Exalytics boxes can be daisy-chained
together using InfinBand connections
‣ For HA scenarios, does not increase
available RAM
‣ Summary advisor scripts write to both TimesTen
databases, replicating aggregates
‣ TimesTen databases can be “wired together”
for failover/HA purposes
TimesTen
Memory-Resident
Database
Checkpoint
Files
Log
Files
ODBC
Oracle BI
Server
nqcmdSummary
Advisor
14Friday, 5 July 13

Summary Advisor for Aggregate Recommendation & Creation
•Utility within Oracle BI Administrator tool that recommends aggregates
•Bases recommendations on usage tracking and summary statistics data
•Captured based on past activity
•Runs an iterative algorithm that searches,
each iteration, for the best aggregate
•Could we use this to cache commonly-used BI Apps
aggregations in TimesTen, automatically based on usage patterns?
15Friday, 5 July 13

Alternative Approach - Copy “Hot Data” into TimesTen for Exalytics
•Standard approach is to store aggregates in the TimesTen datamart
‣ Aggregated by the source DB, aggregates then cached in TT database
•Other approaches could be used, however
‣ Store whole detail-level dataset in the TT database
‣ Store just recent detail-level data in TT, and use OBIEE’s fragmentation feature
‣ Store aggregate layer from BI Apps DW entirely in TimesTen
•Would this be an option that we could use with BI Apps datasets?
16Friday, 5 July 13

Another Option - Oracle In-Memory Database Cache (IMDB)
•Automatically replicate “hot” transactional data from Oracle BI Apps DW tables into TimesTen for Exalytics
•Use OBIEE fragmentation to enable automatic navigation between sources
•Aggregation performed by both TimesTen,
and by source DB (as appropriate)
•However - fairly intrusive approach, Oracle-only,
probably not attractive to most
BI Apps customers and DBAs
17Friday, 5 July 13

Running BI Apps In-Memory - Is it Possible and Practical?
•Can we use the Summary Advisor to automatically cache commonly-used aggregates in-memory?
‣ Similar to regular OBIEE caching, relies on query repeatability + use of aggregation
•Could we copy all, or part, of the BI Apps data warehouse directly into TimesTen?
•How would we update the RPD to point to the in-memory tables?
•How fast would TimesTen be to load, and to query, vs. Oracle/SQL Server/DB2 etc?
•Here’s our thoughts and R&D to date....
18Friday, 5 July 13

Using the Summary Advisor on BI Apps 7.9.6.4 Step 1 : Start Advisor
•Server has to be an Exalytics server, in this example is patched-up to 11.1.1.6.9
•Workstation has the 11.1.1.6.9 BI Administration tool installed
•Select Tools > Utilities, then Oracle BI Summary Advisor from utility list
1
2
19Friday, 5 July 13

Using the Summary Advisor on BI Apps 7.9.6.4 Step 2 : Scope Source Queries
•By default, all queries registered in the usage tracking and summary statistics tables are in-scope
•Refine the recommendations by limiting timeframe, and setting minimum accumulated time threshold
•Still an opportunity later on to pick and choose from recommended aggregates
•Once selected, then select the TimesTen connection pool and database as the aggregate table target
3
4
20Friday, 5 July 13

Using the Summary Advisor on BI Apps 7.9.6.4 Step 3 : Create Aggregates
•Summary Advisor then recommends a set of “candidate” aggregates, which you can choose to implement
•Select all, none or some of the recommended aggregates
•Then run the resulting logical SQL script using the nqcmd utility
•Note - may need to clean-up BI Apps DW data to remove duplicates etc before script completes OK
21Friday, 5 July 13

Using the Summary Advisor on BI Apps 7.9.6.4 Step 4 : Review RPD and Data
•Aggregate Persistence process called by the “create aggregates” process also maps tables in RPD
•Physical layer contains entries for the TimesTen tables
•Business Model and Mapping later contains vertically-federated LTSs for the new TT tables
22Friday, 5 July 13

Summary Advisor for BI Apps : Pros and Cons
•Pros
‣ Automatically analyzes query patterns and recommends aggregates to accelerate them
‣ Handles the registering of the TimesTen tables in the RPD, including mapping back into business model
‣ Supports any data source that the BI Server supports
•Cons
‣ Queries have to have run before they’ll be considered for loading
into TimesTen for Exalytics
‣ Relies on subsequent queries being able to use those aggregates
‣ Could get unwieldy if many aggregates are registered in the RPD
‣ Summary Advisor process does not automatically clear down
tables that don’t feature in future recommendations
‣ Inefficient refresh process, unless you use a process such as
http://www.rittmanmead.com/2013/04/incremental
-refresh-of-exalytics-aggregates-using-
native-bi-server-capabilities/
23Friday, 5 July 13

Alternative Approach - Copy “Hot Data” into TimesTen for Exalytics
•Standard approach is to store aggregates in the TimesTen datamart
‣ Aggregated by the source DB, aggregates then cached in TT database
•Other approaches could be used, however
‣ Store whole detail-level dataset in the TT database
‣ Store just recent detail-level data in TT, and use OBIEE’s fragmentation feature
‣ Store aggregate layer from BI Apps DW entirely in TimesTen
•Would this be an option that we could use with BI Apps datasets?
24Friday, 5 July 13

Loading data directly from Oracle into TimesTen with ttLoadFromOracle
•The procedure ttLoadFromOracle uses OCI to load data directly
from Oracle into an existing TimesTen table
•Specify a whole table (SELECT *), or part (SELECT ... WHERE)
•Target table must existing on TimesTen already
‣ Create it automagically using ttTableSchemaFromOraQueryGet
or createandloadfromoraquery
‣ However both these use Oracle data types and no compression,
so size in memory is going to be greater
REVENUE_F_TS
REVENUE_F_LARGE
ttIsql --ConnStr
"DSN=BISAMPLE_TT;UID=SH;PWD=SH;OracleNetServiceName=
orcl;OraclePWD=SH"
Command> call ttLoadFromOracle('A_TEST',
'REVENUE_F_TS', 'SELECT SHIPTO_ADDR_KEY, OFFICE_KEY,
EMPL_KEY, PROD_KEY, ORDER_KEY, REVENUE, UNITS,
DISCNT_VALUE, BILL_MTH_KEY, BILL_QTR_KEY,
BILL_DAY_DT, ORDER_DAY_DT, PAID_DAY_DT, DISCNT_RATE,
ORDER_STATUS, CURRENCY, ORDER_TYPE, CUST_KEY,
SHIP_DAY_DT, COST_FIXED, COST_VARIABLE,
SRC_ORDER_NUMBER, ORDER_NUMBER FROM
BISAMPLE.SAMP_REVENUE_F_LARGE WHERE
BILL_MTH_KEY=201012');
< 7750 >
1 row found.
-- this has loaded 7750 rows for a given month
TimesTen is
loaded from the
results of a query
on Oracle
ttLoadFromOracle
25Friday, 5 July 13

Generating TimesTen load DDL and scripts with ttImportFromOracle
•For ttLoadFromOracle to work, the target table must exist
•The utility ttImportFromOracle is useful here.
‣ It can map Oracle data types to optimal TimesTen ones
‣ Optionally, it can aggressively limit column sizes based on data to reduce TimesTen footprint
‣ It can evaluate compression effectiveness and apply it only where most useful
‣ Given a set of tables, it will generate:
- TimesTen DDL for requires schemas/tables/indexes
- A script to load all the tables into TimesTen in parallel (ttPDL.sh)
CreateIndexes.sql
ttImportFromOracle
REVENUE_F
CreateTables.sql
CreateUsers.sql
DropIndexes.sql
DropTables.sql
LoadData.sql TableList.txt
ttPDL.sh ttSizing.sh
UpdateStats.sql
26Friday, 5 July 13

•ttImportFromOracle is new in TimesTen 11.2.2.5
‣ Not an official production utility - best efforts support only
‣ But it’s only generating scripts, which contain standard (supported) TimesTen functionality
•The executable is located in $TT_HOME/support
•It uses OCI, so make sure LD_LIBRARY_PATH is set to include Oracle DB lib
‣ export LD_LIBRARY_PATH=$ORACLE_HOME/lib
•Feature-rich syntax, but at its simplest can just be invoked for a single table, with compression:
$ ttImportFromOracle -oraconn SH/SH@orcl -tables REVENUE_F_TS -compression 1
Beginning processing
Resolving any tablename wildcards
Eliminating any duplicate tables
Getting metadata from source
Generating database user list
Assigning TimesTen datatypes
Analyzing source tables
Analyzing table 'SH.REVENUE_F_TS' ...
Estimating table sizes
Evaluating parallel data load
Generating output files
Finished processing
27Friday, 5 July 13

•ttImportFromOracle writes a set of scripts that are subsequently executed to :
‣ Create the target tables and indexes on TimesTen, using optimised data
types and compression
‣ Load the target tables on TimesTen, still via ttLoadFromOracle
REVENUE_F
ttPDL.sh
REVENUE_F_TS
ttLoadFromOracle
REVENUE_F_TS
CREATE TABLE ...CreateTables.sql
28Friday, 5 July 13

Comparing Oracle and TimesTen as data sources
•Our testing has shown that in general,
‣ For base level data, Oracle outperforms TimesTen
‣ For aggregated data, TimesTen outperforms Oracle
•Therefore entire lift + shift of OBIA Data Warehouse into TimesTen is possibly not going to give optimal
response times
-- same query over OBIA
-- 0.68 seconds
Select * from samp_revenue_f_large f, samp_customers_d
cd, samp_addresses_d ad, samp_products_d pd
where cd.cust_key = f.Cust_key and cd.address_key =
ad.address_key
and pd.prod_key = f.prod_key
and ad.city = 'San Francisco'
and pd.type = 'Cell Phones'
and f.bill_mth_key = 201012
;
-- using full sized fact table (native data types)
-- 0.78 seconds query time
Select * from revenue_f_native f, customers_d cd,
addresses_cd ad, products_d pd
ad.address_key
and f.bill_mth_key=201012
;
0.68 seconds 0.78 seconds
29Friday, 5 July 13

Comparing Oracle and TimesTen as data sources
•One option would be to put just “Hot Data” (eg current month) into
TimesTen, and then update the RPD to use fragmentation
‣ This has an overhead in terms of RPD updates (and support - added
complexity), as well as an additional “ETL” process to manage
-- using full sized fact table (native data types)
-- 0.78 seconds query time
Select * from revenue_f_native f, customers_d cd,
addresses_cd ad, products_d pd
ad.address_key
and f.bill_mth_key=201012; 0.78 seconds
-- using Time slice table
-- 0.25 seconds
Select * from revenue_f_ts f, customers_d cd, addresses_cd
ad, products_d pd
ad.address_key
and f.bill_mth_key=201012; 0.25 seconds
30Friday, 5 July 13

Conclusions
•The scale of most BI Apps implementations means that query performance can be an issue
•Exalytics’ TimesTen In-Memory Database could be a potential solution to this issue
•Several approaches to putting all or part of the BI Apps DW into TimesTen / In-Memory
•Copying all or sections of the BI Apps DW into TimesTen is a potential approach
‣ But current version of TimesTen best suited to smaller tables and datasets
•Exalytics’ Summary Advisor now works with Oracle BI Apps 7.9.6.4
•Automatically detects and recommends suitable aggregates, builds and maps into RPD
‣ Though custom solutions are probably more efficient for their later incremental refresh
•A work in progress - speak to Rittman Mead for more details on how this can work
•Offers the potential of “speed-of-thought” business analytics dashboards, with minimum additional work
31Friday, 5 July 13

Thank You for Attending!
• Thank you for attending this presentation, and more information can be found at http://www.rittmanmead.com
• Contact us at info@rittmanmead.com or mark.rittman@rittmanmead.com
• Look out for our book, “Oracle Business Intelligence Developers Guide” out now!
• Follow-us on Twitter (@rittmanmead) or Facebook (facebook.com/rittmanmead)
32Friday, 5 July 13

In-Memory Analytics with Oracle BI Apps and Oracle Exalytics
UKOUG Analytics Event, London, July 2013
Mark Rittman, Technical Director, Rittman Mead
33Friday, 5 July 13

In-Memory Oracle BI Applications (UKOUG Analytics Event, July 2013)

Recommended

More Related Content

What's hot (20)

Similar to In-Memory Oracle BI Applications (UKOUG Analytics Event, July 2013) (20)

More from Mark Rittman (14)

Recently uploaded (20)

In-Memory Oracle BI Applications (UKOUG Analytics Event, July 2013)