Export OLAP 11G AW/Cubes to OBIEE 11g

Hi Experts,
If anyone has handled this assignment of exporting OLAP 11G AW/Cubes to OBIEE 11g, Please share all details. Currently we use BO for reporting requirements and before going ahead with OBIEE 11g would like to make sure there is not much of a gap. As of now we see a lot of incompatibility while exporting like some dims and cubes cannot be exported and so on. Any white paper/docs/information on this from experienced folks are welcome.
We are on 11.2.0.3 db release.
Thanks in advance,
MS

Here is some basic information about importing 11g analytic workspace cube and dimension metadata into OBIEE.
There are two basic options as I suspect you already know:
Option 1) Before the release of OBIEE 11.1.1.5 (I believe), there was no native support for the OLAP option within the administrator tool of OBIEE, and
one needed to use the AwM plug-in to OBIEE. The AwM plug-in generates a UDML file which can be imported into OBIEE using the OBIEE admin tool.
The AwM plug-in to OBIEE generates metadata for the source, business model, and presentation layers. The very latest version of the AwM plug-in will specially
handle calculated measures created in AwM and create the proper calculcated measure expression in OBIEE. This is very important if you are querying above the cube
to get the correct aggregation results. One area that has limited support is if a dimension only has value-based hierarchies. The plug-in will create a very simple level-based
hierarchy to at least allow the dimension to be successfully created. You then need to do more work within the OBIEE administration tool to create the fully baked
value-based hierarchies. I do not have any real-world experience with value-based hierarchies in OBIEE. Again, the AwM plug-in to OBIEE works with level-based hierarchies
and supports multiple hierarchies per dimension. The AwM plug-in to OBIEE is available on the OTN download page for the OLAP option.
Here is the link: Oracle OLAP Downloads: Software & Samples
Option 2) OBIEE 11.1.1.5 and later allows for native integration for accepting OLAP 11g dimension and cube metadata as a supported source.
This integration makes the AwM plug-in somewhat obsolete (except for the calculated measure expression integration which is unique to the AwM plug-in and using
AwM to create calculated measures including time-based calculated measures)
Please expand on the incompatibility issues you are running into. Is this related to a lot of value-based hierarchies or other issues?

Similar Messages

Building OLAP 11g Cubes Tutorial

I followed the tutorial [http://www.oracle.com/technetwork/database/options/olap/buildicubes-090657.html#o]
all the way to the section of Mapping cubes to relational sources and I got stuck. Since the images / diagrams do not come up I was unable to figure it out, can somebody assist me on this matter. Maybe there is a way to be able to view the images / diagrams. I am very new to this cubes and so far I find it very interesting, but I still want to understand fully the advantages of using cubes in reporting in OBIEE.
http://www.oracle.com/technetwork/database/options/olap/buildicubes-090657.html#o
Thanks

Not sure why the images are not showing up on this website. Maybe someone from development can get it fixed.
Regarding benefits of using OLAP cubes with OBIEE, see recent posts:
HOLAP - using OBIEE and Oracle OLAP
and
Usage of OLAP
There are lot of benefits. It simplifies reporting, as the calculations and aggregations are "pushed" inside the database (i.e., OLAP cubes).
No need to create MVs or aggregate tables.
Adhoc Query performance is excellent, no matter what level of data you are looking at.
Calculations are very easy to create, and automatically works at all levels of all dimensions.
OBIEE rpd work is extremely simple, and very minimal - compared to obiee reporting from relational sources.

OLAP 11g BUILD CLEAR example

Can anyone please help me find the syntax/plsql block of clearing dimension data from cube in OLAP 11g?
Thanks

BEGIN
DBMS_CUBE.BUILD('OLAPTRAIN.CHANNEL USING (CLEAR)');
END;
Note that when you clear a dimension, you also clear data from every cube that is dimensioned by that dimension.

Are there any Oracle OLAP ODI Knowledge Modules for Oracle Olap 11g ?

Hi guys, wasnt sure where to post this so I put it here and the ODI forum
Does anybody know when Oracle OLAP Knowledge Modules, will be available for Oracle Olap 11g. It appears that the knowledge modules released with Oracle Data Integrator 10.1.3.5, currently only work with Oracle Olap 10g...
If there is no timeline, does anybody know of work around to get ODI reverses working on Analytical Workspaces built in Oracle Olap 11g
Thank you
Richard
Edited by: RichardSmith on May 15, 2009 11:51 AM

Raise a service request with Oracle Support (Metalink) for your system slow issue. May be, they will recommend you the workarounds for the issues which are not yet available for 11g in the form of Patches.
Regards,
Sabdar Syed.

Importing the dump into 10g which exported from 11g server

hi
can i import the dump into 10g which exported from 11g server?
thanks
nidhi

knidhi wrote:
hi
can i import the dump into 10g which exported from 11g server?
thanks
nidhiHi Nidhi,
No, if you want to use an 11g export dump and import it using 10g utilities to a 10g database.
Yes, if you use 10g export utility to extract 11g database data, and use 10g import utility to import in 10g database.
For more information on export/import compatibility, please check:
http://download.oracle.com/docs/cd/B28359_01/server.111/b28319/exp_imp.htm#
HTH
Regards,
Z.K.

Export from 11g and import to 9i

I have run exp from machine runing oracle 9i to export tables from
another machine running oracle 11g .Export is working fine but while
importing in Oracle 9i it is giving the following errors.
import done in US7ASCII character set and AL16UTF16 NCHAR character
set
import server uses WE8ISO8859P1 character set (possible charset
conversion)
IMP-00003: ORACLE error 96 encountered
ORA-00096: invalid value for parameter plsql_compiler_flags, must be
from among NON_DEBUG, DEBUG, INTERPRETED, NATIVE
IMP-00000: Import terminated unsuccessfully
Can anybody suggest how to do this.
Thanks for your valuable time.

Duplicate post
export from 11g and import to 9i
Srini

Import a cube in obiee

Hi,
my config:
windows Vista
obiee 10.1.3
sql server 2008
Followed this [blog |http://www.rittmanmead.com/2008/02/01/obiee-hybrid-olap-reporting-using-ms-analysis-services-oracle/] to be able to import analys services 2008 cube in obiee. But when try to import a cube in obiee i get the following message: nqserror 46100: sax parser failed with error expected whitespaces
what i exactly did:
1) download en installed the XML for Analysis SDK
SSI:
1) i created a connection pool --> named it xmla and click ok
2) add a new virtual directory --> alias: xmla , physical path: C:\Program Files\Microsoft XML For Analysis SDK\isapi
3) right click on my virtual directory --> convert to application : set the application pool to the one i just created.
4) double click "handler mapping" --> add script map --> request map = *.dll, excutable msxisapi.dll, name = xmla, do you wat to allow this asapi extension --> i clicked yes
5) authentication --> ananomous is enabled by default
obiee:
import --> analys service 2000 --> url i used http://localhost/xmla/msxisapi.dll
import --> analys service 2005 --> url i used http://localhost/xmla/msxisapi.dll
When trying things out i also set the physical path credentials: username/pass. But still doesn't work...
Hope some can guide me throw a good working obiee import or give me some new ideas.... tnx in advanced....
UPDATE:
I've got the iis working oke now. The error above has to something, that the URl i reference to could not be found. Now when testing IIS I get the xml back in my browser. When trying in in obiee, it gives me the error:
nqserror 64203 XML/A eror returned from the server. Fault code: "XMLAnalysisError.80004005"Fault string "an error occured while processing the request. :
Edited by: user565199 on 18-nov-2009 5:13

Please, can anaybody help? tnx in advanced...

Connecting SQL Analysis Services 2005 cube with obiee 10g

Hello experts,
I'm trying to link an SQL Analysis Services 2005 cube with obiee 10g.
In Oracle BI Administration, when I choose File -> Import -> From
Multi-dimensional, I have to fill the next gaps:
Provider Type: Analysis Services 2005
URL: ***********
User: blank
Password: blank
Before to do that, I need to create the URL path.
To create the URL, I have followed this link.
http://erpthings.blogspot.com/2008/08/obiee-connection-to-msas-2005-
cube.html
But, it hasn't worked because when I was trying to configure the HTTP
access to SQL Server 2005 Analysis Services, we have to perform a test to ensure that it has been well configured. In IIS -> WebSites -> Default WebSites -> olap, we should browse the document called "msmdpump.dll" and this should show an xml document, but it doesn't.
Thanks!

Hi Alex
It sounds like your MS data pump might not be configured correctly. First ensure that is working before trying to put the values into OBIEE. Don't worry about the 500 error - I get that too instead of an XML page.
You can read a bit more about SSAS support on my blog here:
http://total-bi.com/2010/12/obiee-sql-server-analysis-services-cubes/
Here's the link to Microsoft's description of setting up msmdpump.dll
http://technet.microsoft.com/en-us/library/cc917711.aspx
Paul

How to Export OLAP analytical workspaces

Hi, experts:
I am following Doc. Migrating OLAP From 32 Bits to 64 Bits or Across Platforms (Doc ID 352306.1) to export OLAP analytical workspaces.
I copied and pasted the step 2. as below. But I donot understand some of them. Hope you can help me.
1. What does this mean by: "(NOTE: this following statement is critical so everything in the AW is in STATUS)"?
2. "Replace ALIAS_DIR with a defined directory alias"   This ALIAS_DIR can be anything on current directory?
3. I have more 3 AW. So I have to run the following four scripts 3 times? I have to modify the export_file_name.eif file name for each run, right?
========
2.Use the following SQL commands, for EACH AW that was identified in step #1   (NOTE: All quotes are single quotes). Caution: If you attach multiple AWs and then run the   export command, ONLY the last- attached AW will be exported. --- Replace OWNER.AW_NAME with the the actual OWNER.AW_NAME name SQL> exec dbms_aw.execute('aw attach OWNER.AW_NAME rw'); -- (NOTE: this following statement is critical so everything in the AW is in STATUS) SQL> exec dbms_aw.execute('allstat'); --- Replace ALIAS_DIR with a defined directory alias --- Replace export_file_name.eif with a valid eif file name SQL> exec dbms_aw.execute('export all to eif file ''ALIAS_DIR/export_file_name.eif '''); --- Replace OWNER.AW_NAME with the actual OWNER.AW_NAME name SQL> exec dbms_aw.execute('aw detach OWNER.AW_NAME');
=======
Thank you very much for your help.

1. What does this mean by: "(NOTE: this following statement is critical so everything in the AW is in STATUS)"?
This is just a note to the readers and you can simply run the sql statement as it is in the doc.
2. "Replace ALIAS_DIR with a defined directory alias"   This ALIAS_DIR can be anything on current directory?
Correct but you need to create the directory first -- See Error - EBS DB importing OLAP from 32 bit to 64 for the commands.
3. I have more 3 AW. So I have to run the following four scripts 3 times? I have modify export_file_name.eif file name for each run, right?
Correct.
Thanks,
Hussein

Integrating Cognos Cubes into OBIEE

Hello,
Is there any way to integrate cognos cubes+ into OBIEE and use them in OBIEE..?
Thanks in Advance..

No direct one for the time being. You can't import Cognos metadata into the rpd like you would for Essabse or MS Analysis Services.
Check this post Cognos to OBIEE
Cheers,
C.

Accessing OLAP 11g Cubes

Hi Experts,
We are on OLAP 11.2.0.1 and have built cubes using AWM 11g. What are the different and best ways to provide access to these OLAP cubes for reporting. We are looking at Simba MDX provder. Any idea if XML Publisher can be used to provide access to these cubes? If so, Can you please let me know from where can i get any information/documentation related to this? We would ideally want to study the pros and cons b/w the various ways of access 11g OLAP cubes.
Look forward to suitable replies.Thanks in advance.

You can refer to this page here => http://www.oracle.com/technetwork/database/options/olap/index.html
You also have APEX to query the olap cubes other than excel addin from simba and OBIEE.
Thanks,
Brijesh

Error while maintaing olap 11g cube

Hi,
Iam getting the below error while trying to maintain the cube
An error has occurred on the server
Error class: Express Failure
Server error descriptions:
INI: Error creating a definition manager, Generic at TxsOqConnection::generic<BuildProcess>
INI: XOQ-01724: Can not specify a USING clause with a rewrite MV cube, Generic at TxsOqBuildProcessor::initializeForDimension
at oracle.olapi.data.source.DataProvider.callGeneric(Unknown Source)
at oracle.olapi.data.source.DataProvider.callGeneric(Unknown Source)
at oracle.olapi.data.source.DataProvider.executeBuild(Unknown Source)
at oracle.olap.awm.wizard.awbuild.UBuildWizardHelper$1.construct(Unknown Source)
at oracle.olap.awm.ui.SwingWorker$2.run(Unknown Source)
at java.lang.Thread.run(Unknown Source)
ALso when I try to drop the cube/dimension , I get the following error
An error has occurred on the server
Error class: Express Failure
Server error descriptions:
DPR: Unable to create server cursor, Generic at TxsOqDefinitionManager::generic<CommitRoot>
INI: XOQ-00703: Error executing OLAP DML command "(AW ATTACH CBU_UNIFIED.RECHARGES RW : ORA-34738: A severe problem has been detected. Analytic workspace operations have been disabled.
)", Generic at TxsOqAWManager::executeCommand

The two errors are, I suspect, unrelated.
The XOQ-01724 error happens if you attempt to execute dbms_cube with a script that contains a USING clause. This is allowed for most cubes, but not for cubes associated with a materialized view. If this is what you tried to do, then the error is expected and the solution is to not specify a USING clause or to drop the MV.
The second error is much more serious since it means your AW is probably corrupted. Did this happen after a failed parallel build? What version are you using.

How do I know what is the minimum resource requirement for OLAP 11g?

How do I know for OLAP 11gR1 ...
1. what is the minimum number of CPUs and speed required
2. how much minimum physical memory, SGA, PGA etc. is required?
Regards, Anirban

Hello Anirban,
In summary, since the OLAP option is an option to Oracle Database 11g, Enterprise Edition the minimum requirements in order to run the database would be identical to those necessary to run a "Data Warehousing" type installation. That being said starting up a database is a far cry from actually performing well while accessing and building OLAP Cubes in your Database.
There are lots of things to consider. For example: the complexity of your model; data sparcity; level of aggregation; system load, SLAs, etc..
Lets look at true minimum requirements and then consider what it would take to achieve good performance in your environment
TRUE MINIMUM...
I performed a test in my environment (Oracle Database 11g Release 2 (11.2.0.2.0) on Oracle Enterprise Linux 5.5) using Analytic Workspace Manager 11.2.0.2.0A and found the following minimum settings to work against the OLAPTRAIN schema (You may find it here http://www.oracle.com/technetwork/database/options/olap/index.html). This is a very small schema which we use for demonstration purposes.
I discovered that the database started with a memory_target of only 160m (apparently this is the absolute minimum for my environment). I knew that if the database successfully started that I would be able to query the Cube but maintain it? Running queries against the OLAPTRAIN Sales Cube, despite being slow, worked as anticipated. Next, I performed a maintain operation on the Sales Cube where I had to abort the job because it ran into memory issues at 160m, however, when I increased it to only 180m it did a complete maintain in a reasonable timeframe, albeit slow considering the limited resources.
MINIMUM REQUIREMENTS TO ENSURE REASONABLE PERFORMANCE IN YOUR ENVIRONMENT
Most importantly ensure you have a good cube design
Run a sample set of your data in order to define what "reasonable performance" is for your environment. Create a small representative cube in order to test both build and query performance while collecting and analyzing database statistics through AWR and ADDM. For OLAP option specific queries please take a look at http://wiki.oracle.com/page/OLAP+option+-DBASample+Scripts.
You could creat a view on top of your fact table and use the view to generate your OLAP Cube. This will allow you to test a more manageable amount of data. For example, if you have 10 years of data and want to determine the footprint and resources required in order to manage a Cube of this volume then consider scoping it down to a few months first then increase it by a measured amount in order to predict requirements. Monitor your resources and ensure you balance I/O, memory and processors. Since the OLAP option is a Enterprise Edition option it falls under Data Warehousing. As such there is plenty of literature available on oracle.com in order to define what it takes to achieve a balanced Data Warehouse.
I would recommend contacting Oracle Consulting or an Oracle Partner to determine the best fit for your environment.

Eif import/export in 11g

Hi All,
I exported my analytic workspace into an eif file and imported into another analytic workspace. The Analytic workspace that I imported into had already built in dimensions and cubes. Before I import eif file I deleted Analytic workspace using AWM and then imported. Import succeeded without any errors so when I import what happens to the views that AWM created automatically in my imported Analytic workspace. How can I enforce drop old dimension/cube views and create new views from my eif file? because in my case my schema still had old views from old AW objects instead of creating views from eif file.
The problem I noticed was when I tried to refresh cubes using DBMS_CUBE package. I was getting following error as my build referring views from my old cube.
ORA-37162: OLAP error
XOQ-01707: Oracle job "JOB$_1132" failed while executing slave build "TEST_AW.TEST USING (LOAD NO SYNCH, SOLVE) AS OF SCN 2299502303986" with error "37162: ORA-37162: OLAP error
ORA-36810: Analytic workspace object TEST_AW.TEST_AW!MY_OBJECT_ID_L3 does not exist.
ORA-06512: at "SYS.DBMS_CUBE", line 236
ORA-06512: at "SYS.DBMS_CUBE", line 305
ORA-06512: at line 1
Here object TEST_AW.TEST_AW!MY_OBJECT_ID_L3 belongs to old cube which I removed in my new AW. How can I force my build to look at new SCN?
Thanks
Dileep.
Edited by: user3953230 on Feb 28, 2011 3:21 PM

Dileep,
This sounds like
BUG 9761434 - KGL OBJECTS NOT FLUSHED FROM ALL EXADATA NODES DURING AW DELETE OPERATION
The basic problem is that the old definition of the AW gets kept in the 'shared pool', also known as 'kgl', of a RAC node after the AW is deleted. If you immediately create an AW with the same name in another RAC node, then the system can get confused. The workaround is to issue
alter system flush shared_pool;in all RAC nodes.

OLAP 11g CUBE_BUILD_LOG

Hi to all.
In 11g when you load a cube, logging is written in CUBE_BUILD_LOG table. I don't see there any information about the number of rows loaded during a build. Is there any other place that I can find such information?
I need the number of rows just to be sure that all rows from the fact table were loaded in the cube.
Thank you very much.
Chris

Thank you Anne and Stuart.
Stuart yes it would be very helpful to have an example. I'm investigating a specific performance problem and it would be helpful to see actual number of rows that were loaded.
The problem is that although I partition on QUARTER on a time dim, when I load the last month of 1st quarter I have a SOLVE time of 31m (quarter rows including last month=1.5m). When I load the last month of 2nd quarter, the SOLVE time gets up to 57m (almost double with again 1.5m rows for the 2nd quarter). As I beleive, the SOLVE time for each quarter should be identical provided that the rows loaded for each quarter are identical (I use precompute 0% on top and 20% on button). LOAD times are identical for all months.
Anyway, when I conclude the tests I will open a separate thread for this, to be consistent with the subject.
Thank you very much.
Chris.

Export OLAP 11G AW/Cubes to OBIEE 11g

Similar Messages

Maybe you are looking for