Transeparent partition in ASO (synchrinisation)

Hi All,
I have the problem with partition (transeparent) wen im loading the data in one cube its not reflecting exact value in the terget cube ,im clearing the data loading again &again but its showing different value both measures are not same in the cubes (srce& target) ,, wen im loading the changes its not reflecting in the target cube may be its not sychronising so how can i Resynchronise the partition
I think u understing wat i Have said,plz any help would be appriciated
Thanks
Ram
Edited by: R@Mn@NdU on 28 sept. 2009 07:46

Hi,
We do not do synchronization in Transparent partitions. The updated values in source should automatically reflect in the target.
I suspect that you are not bringing all dimension members in the partition. If measures are not same in source and target, define only the measures in partition, which are same in both source and target. Also, check your other dimension members defined in partition, in addition to measures
if this does not help, I need more information about your dimensions and the members you are partitioning on.
Cheers
RS

Similar Messages

Retrieval very Slow when using Partition in ASO

Hi - I have partitioned my ASO cube using Transparent partition and I amretrieving it thru BSO.The retrieval for some combinations are VERYVERY SLOW in Excel.We have over 250,000 members.I have usedRetrieval Buffer and Retrieval Sort Buffer sizes of 10000KB and1000KB respectively.Some retrievals take more than 4 minutes. When Designing Aggregation for ASO, I have used the Recommendedviews to materialize aggregation. Do I have to do or change some settings for better Performance? Any help would be greatly appreciated. Thanks, Narain

Making direct I/O (in place of default buffered I/O) candefinitely help a lot.Then you can increase some data file cache upto 2GB [as per yoursys support]. Also check about your dedicated parallel processsides.I wish it helps.

Transparent Partition Between ASO & BSO

Hi,
I am trying to define a Transparent Partition between an ASO (Source) & a BSO (Target) cube.
My ASO cube has 16 dimensions, BSO has 12 dimensions. The matching 12 dimensions are identical.
The data from the source should be taken from all levels of the 12 matching dimensions and Highest level data of the non matching dimensions.
Is this doable?
Is there any special syntax in partitions definitions?
Appreciate your thoughts.
Thanks,
Ethan.

Glenn,
Thanks. It's working. I created the partition manually, kept void for all the 4 missing dimensions.
When I have to write the script, I am not sure how to specify that in the script
mapped targetAreaA ("Dim1") to (Void)
mapped targetAreaA ("Dim2") to (Void)
mapped targetAreaA ("Dim3") to (Void)
mapped targetAreaA ("Dim4") to (Void);
or
mapped targetAreaA ("Dim1", "Dim2", "Dim3", "Dim4") to (Void, Void, Void, Void)
Are these correct syntaxes?
Appreciate your thoughts.
Thanks,
Ethan.

Partitioning from ASO to ASO

Hi ,
Basically we have a reporting cube (Assume like A) and other cube which holds GL data and adjustments (assume like cube B) both of these are ASO.
As a part of daily refresh of Cube A (reporting cube). The level zero data is exported and loaded back . And GL data is loaded from a file. Apart from this write back are exported from cube B and then loaded to cube A.generally we maintain three years of write backs in reporting cube.this write back export and import take place as a part of refresh.
So i created a transparent partition between both cubes source as B and target as A. Things are working fine but problem is like the account heirarchy n some other dimension structures are different in both the cubes so i collected all the members of dimensions and included in the mapping region of partition rather than using (@children,@descendants n all those) as heirarchy is not the same.
. And the problem here is , business keep adding members under the dimensions. And i cant manually make the changes to include those new members in partition area. Is there any way of dynamically adding newly added members in the partition area.

What is the problem if you mention @IDESC("Account") in Source and destination Mapping . That way everything is Dynamic . I have a dimension where members keep getting added , I use this simple technique which works well.

ASO to BSO Transparent Partition problem

Hi,
I am doing a partition from ASO(Source) to BSO(Target). Outlines are same except one dimension:
Aso cube
dimension has 10 Level 0 members in Level 1.
BSO cube
Same dimension has 2 level 0 members(these 2 members are same as in ASo) in Level 1.
Do rest 8 members be available while retrieving data in BSo cube? [Right now when i connect can only see 2 members]
Was wondering with Transparent partition is it possible to see the additional outline members in BS ?.
Any help to clarify would help thanks.

I think i am understanding your question.
A user will only "see" what members/dimensions are in the Target cube. You would need to add the other 8 members to the target outline to be able to pull the data for them from the source cube.

Convert ASO outline to BSO

I am building the outline in ASO using rules files and MAXL script...How can I convert this outline in BSO using the same rules files...or is there any other way to convert the outline to from ASO to BSO.... I need this for creating a ASO-BSO transparent partition where ASO is the data source and BSO is the target....
Thanks...

In your case.
1. Create a BSO application, and use the same rule files and scripts which you are using for ASO outline build.
2. Within the BSO application,validate whether the same rule files are correct .
3. Mostly ,they should work,else , the minimal changes might require to make them compatible with BSO nature.
4. Now that you have both BSO and ASO cube in place, make partitions.
You might face problems when you do the rule files in BSO ,then you can post them and get corrected.
Sandeep Reddy Enti
HCC

Transparent Partition Issue

Hi All,
We have Planning BSO cube and it has 16 dimensions (Had to go with that many dim, although the max. is 10), and the aggregation process took abt 1.30 hrs, we've tried all possible fine tuning.. considering the no. of dims, it's fairly ok.
Now the client wants all the HFR reports to be in real - time, so we've created a reporting cube (ASO) with exactly same dimensions and members, created a transparent partition mapping only level 0 members, and expected it work like once the level 0 members are mapped, it'll automatically aggregate to the parent members since its "ASO".
Can you please let me know whether this logic will work or I've to go to replicated partition?
Forgive my ignorance if my judgement is completely wrong.
Regards,
Rajeevan D
Edited by: 865958 on Jun 14, 2011 10:54 PM

Your question jumped from what sounded like data not being replicated in the partition to the outlines getting out of sync. I'm guess the latter is more likely a problem.
Is the partition on ASO? BSO -> ASO partitions need to be manually synchronized.
Regards,
Robb Salzmann

PARTITION WIZARD

'Version 11 partition wizard does not allow you to partition 2 ASO cubes (still looks for block storage). If you want to create a transparent partition between 2 ASO cubes, do you have to do it manually, or is there a way of tricking the wizard?

I agree that funky partition schemes for Disk 0 is not recommended. Nevertheless it is not rare for our IT dep. to re-install PC's with dual boot configurations and/or data partitions on Disk 0. So instead of backing up everything prior to re-installing
Windows, it would come in handy to be able to dynamically choose which partition to install Windows on.
I really miss a more elaborate guide on how to implement that wizard page, you gave me a link to, but it really does not seem to be what I need after all.
Do you have any comments on this idea: I have added DaRT to my Windows PE. Using the File Explorer I get access to the partitions and their names, enabling me to rename whatever partition (i.e. "WinOS"). What if I made separate Task Sequence(s)
and added a check for which partition with a given name ("WinOS") to install Windows on? Then I could use File Explorer, rename the partition I want to install Windows on ("WinOS"), and thus keeping everything intact except that partition?

Regarding ess0001.dat file

Hello All,
I am assuming that ess0001.dat file in the Application-> default folder is the one which stores all the data in an Application. Just in case I want to replicate that data in a different App , will copying that .dat fi einto the default location of new App makes both Apps have same data ? I am talking about ASO apps by the way.

Well , i did everything except copying App -> Metadata file. May be i will try that next time.
Between , thew source for this question is another Issue I came through last week. I wanted a particular ASO application to be restored to a certain Point of Time , while restoring ess0001.dat file did not restore properly. Got the prompt that file is in use even though I had stopped all the Applications . Nevertheless , there was an option given "Restore on Reboot".
I was reluctant to choose this option , but I eventually did. Later when the Reboot was actually done ess0001.dat file actually went missing in the Application folder both in App-> default and App->Metadata and I could not open ASO application nor the App getting Transparent partitioned to ASO . Then I again had to do a restore , this time luckily did not get Restore on reboot. Just wondering , when previously I had selected "Restore on Reboot" Instead of restoring , it deleted the file itself Hmmmmm . One more thing , Reboot was done to the APS server while I did my operations on Essbase server . So , the Restore on reboot should have been justified when Reboot was done to Essbase server correct , but it happened after rebooting APS. Totally out of Sync.
Any comments John , Team !!

Problem with BSO - ASO partitions (Essbase 9.3.1)

Hi there,
Is it possible to create a BSO partition on top of four ASO cubes when using member set functions in the partition definition (e.g. @GENMBRS or @LEVMBRS)?
I have tried that but always get an “overlapping partitions” error, although the partition members are not defined overlapped. The cellcount does match.
Appreciate any hints.
Thanks in advance.

Obviously this won't resolve your entire problem, but just to answer the first question you can definitely use e.g. @GENMBRS successfully in the partition definitions for a single BSO cube over multiple ASO cubes in 9.3.1.

ASO to ASO Partitioning not working in 11.1.1.3

Hi All,
We are to create a transparent partition using EAS between 2 ASO cubes. When we use member relation functions such as @IDESCENDANTS on the dimension name, the partition validates successfully however no data comes across in the target database. the member names match exactly between source and target dimensions. When we change the partition definition to use UDA function on dimension name and level 0 of dimension, the data comes through which suggests the aggregation is occuring dynamically. Not ideal as we have a over 600 users that will retrieve on the target cube.
We have tried to overcome this by building the dimensions in the source database to have Parent members to have the ~ property and level 0 members to be store. This fails as no data comes across.
As you may figure out, we have competing issues of data pull through from source database and retrieval performance.
Any thoughts on what may be going wrong here?
Ryan

Hi Thomas,
I too tried this a few days back. I am using Jdev 11.1.1.6.
Things that I noticed are :
When I only change the duration of the task, then the end time is returned to me using the dataChangeListener.
But, when I drag and drop the task to some other place, then the dataChangeListener returns me the new start time and a null value for the end time. Not sure why it returns me a null value for the end time.
I created one thread for this but no solution as of now. ADF Gantt Chart getEndTime returns null
As a workaround, what I am thinking is when I drag and drop the task, then i'll pass the new start time to a database procedure. Before updating the values in the table, we can get the duration of the task using the old start and end time. Then we can add this duration to the new start time to get the new end time. Now, we will have both the values and the same can be updated in the tables too.
I still have to do the above step. Will let you know if it works fine.
Thanks,
Ankit

Essbase 11.1.2.3 BSO to ASO Partitions with extra dims

Hello Essbase brain trust.
I have a situation where my BSO/planning cube has less dimensions than my ASO reporting cube. I have tried all of the ways mentioned in the documentation to map the extra dimension to no avail. Seems like the documentation is lacking.
Does anyone have a working situation where they could share the exported partition? Or enlightened me.
I did this using the planning mapping tool but it's very slow to refresh although it works. Even the xml from this works not helpful.
Thanks
Dave

It works fine. I tried with ASOSamp and Sample
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
<EASFrameworkTransferObject>
 <Partition>
 <transparent comment="" direction="to" outline="reverse" replace="yes">
 <sourcedb application="Sample" database="Basic" password="bCw6vmd7Zyvt7ZGiTqaotQ==" server="EPM11123:1423" usernme="admin@Native Directory"/>
 <targetdb application="ASOsamp" database="Sample" password="bCw6vmd7Zyvt7ZGiTqaotQ==" server="EPM11123:1423" usernme="admin@Native Directory"/>
 <areaspec sourcealias="sourcearea1" sourcearea=""Jan","Sales","100-10","New York","Actual"" targetalias="targetarea1" targetarea=""Jan","Price Paid","Curr Year","Sale","Cash","Coupon","1 to 13 Years","Under 20,000","Displays","044862","10006""/>
 <globalmap sourcemember=""Jan"" targetmember=""Jan""/>
 <globalmap sourcemember=""Sales"" targetmember=""Price Paid""/>
 <globalmap sourcemember=""100-10"" targetmember=""Displays""/>
 <globalmap sourcemember=""New York"" targetmember=""10006""/>
 <globalmap sourcemember="Actual" targetmember="Sale"/>
 <globalmap sourcemember="" targetmember="Cash"/>
 <globalmap sourcemember="" targetmember="Coupon"/>
 <globalmap sourcemember="" targetmember=""1 to 13 Years""/>
 <globalmap sourcemember="" targetmember=""Under 20,000""/>
 <globalmap sourcemember="" targetmember=""Curr Year""/>
 <globalmap sourcemember="" targetmember="044862"/>
 <globalmap sourcemember="" targetmember=""Store Manager""/>
 <globalmap sourcemember="" targetmember=""Square Footage""/>
 <globalmap sourcemember="" targetmember=""Area Code""/>
 </transparent>
 </Partition>
</EASFrameworkTransferObject>Planning mapping tool uses a different method. It uses (creates on the fly) DATAEXPORT calc script and a load rule to load data.
Regards
Celvin
http://www.orahyplabs.com

Partition in Essbase ASO

Hi Experts,
We have ASO cube with 10 years of data.
The requirement is have two ASO cube with 5 years each and all the 10 years of data should be accessed at the target cube.
Can you please advice how we can do this.
If I select the transparent partition at the target cube it shows only the source data, target cube data is overwriiten by the source cube data.
Thanks in advance for your help guys.
Raja Mohamed

Hello,
You can create 2 news ASO cubes ( don't destroy your ASO cube with all 10 years data). On each new cube you can create a "replicated" partition (target = new cube with only 5 years, source = historical cube) When you execute the replication, the data are loaded from the source to the target. Execute a replication on the two new cubes (delete the replicated partitions when the 2 aso cubes are loaded)
And create a third empty cube (a proxy cube) with the same structure as the two news cubes but with all years and create 2 transparent partitions which source are the new ASO cubes. The end users will only know the third empty cube.
By this way, you can create another ASO cube with n years and your have juste to registrer a new transparent partition in the proxy cube.
I developped this with 9.3.5 version and with this version, the proxy cube must be BSO not ASO (I don't know in newer version)
Good luck

BSO to ASO Partitioning Help

Can anyone help me with a couple questions in regards to our partitioning strategy from our Block Storage planning cube to our Aggregate storage planning cube in 11.1.2? Our original approach was to use a transparent partition for planning scenario (Budget, Forecast) data from BSO source to ASO target for level-zero only. ASO would handle aggregations. Also loading Actuals data directly into ASO cube from source system. We would like to have additional dimensions not required for Planning to be stored just in the ASO database in order to minimize database size and calc times on the BSO side. Two questions:
1) We've found that the Actuals scenario data can not be queried in the ASO target (return #missing). Is this standard behavior for a transparent partitioned ASO target in 11.1.2?
2) Regardless of partition type (Replicated or Transparent) can we have a different number of dimensions in our BSO and ASO cubes? Again my understanding is no, but wanted to confirm.
Thanks in advance for any help that can be provided.

data that is loaded directly to an ASO cube that is the target of a transparent partition is not visible^^^Isn't this the key? I read that the natively loaded data is overwritten (not really, more like overpartitioned) by the partition definition and the partition takes precedence.
So the answer then would be to not have that conflict.
Unless Tom was talking about something else/I'm misunderstanding.
Regards,
Cameron Lackpour
Edited by: CL on Mar 10, 2011 8:13 AM
Whoops, missed this part:
This is the case even when the intersection of the data is outside the area defined by the partition.I still wonder if there's an overwrite of intersections going on.

Partitioning BSO to ASO to get around FDM limitations

FDM only currently loads to BSO Essbase cubes. However, one of our FDM-loaded BSO cube is far better suited to being ASO, not least because it has:
- Twelve dimensions.
- A huge "customer" hierarchy (over 40,000 members), which is purely aggregating.
- Most of the other dimensions are also purely aggregating (although there are a few complex calcs on specific members).
The sheer size of the BSO cube outline and the data being loaded results in prohibitive calculation times, so we are looking for a workaround.
Thoughts at present are to create an ASO cube (effectively copying and converting the current outline to ASO using the Wizard), create a replicated partition to transfer data from the BSO to ASO cube (mostly from level-0 members) and to use the BSO cube purely as a data-entry point. Users would then report from the ASO cube, which could be refreshed from the BSO cube when required.
This would then mean that upper-level members and any calculations in the BSO cube are pretty much redundant.
Does anyone agree with, or have any reservations on the above idea? Is there potentially a better method here? Also, could anyone please hazard a guess on how long the replicated partition refresh might take?

Is the BSO cube only loaded from FDM / other loads? Or are users entering data? Is it an Essbase or Planning app?
While it is true that there isn't a FDM adapter to load into ASO, you do not have to go from a BSO cube to the ASO cube for the data loads. Just have FDM export to a flat file and load that into ASO using maxl / batch script.
Deanna Sunde
Sr. Director
The Hackett Group

Transeparent partition in ASO (synchrinisation)

Similar Messages

Maybe you are looking for