Drill across (joining facts) using Discoverer

Hi,
I've read a few books and article on DW talking about drilling across. It seems a good idea to build multiple star schemas with "conforming" dimensions and then join fact tables (drill across) when needed.
I have tried this solution, and afer a few tests, got to the conclusion that it is preferable to UNION ALL the related facts using a view or a custom folder. Of course this solution is less flexible because we need to find all the required combinations of facts.
Any comments?

OK,
I have 3 dimensions : D1, D2, D3 and 2 facts : F1, F2.
Dimensions D1, D2 are related to fact F1.
Dimensions D1, D2, D2 are related to fact F2.
I want to query on fact F2 based on values of dimensions D1 and D2, and then join F1 to get corresponding measures.
When running the query I get the following error:
Invalid combination of joins using different join keys - query cannot be resolved.
I use Oracle Discoverer 9.0.2.53.12.
What would be the best way to work around this restriction?

Similar Messages

Drilling Across in Discoverer

Discoverer 10.1.2.X
We have datamarts which have uniformly conforming dimensions and the facts and dimensions are laid out in a star schema.
Point of Sale(POS) is a separate fact table and Shipments is a separate fact table too. Stores belong to the customer dimension and Dates belong to the calendar dimension. Both POS and Shipments share the stores and calendar dimension. There is appropriate foreign key relation ship between the fact and dimension tables in the database too.
POS
\|/
|
Stores
|
/|\
Shipments
POS
\|/
|
Calendar
|
/|\
Shipments
The problem:
Now I would like to Drill across in Discoverer i.e. for a given store and date range I would like to get two measures from different facts -> POS and Shipments by day. When I execute the report from Discoverer, the query that is generated looks like the one pasted below and I don't think this is a proper query because cus_dimn_customer and day_dimn_cal_hdl do not have any relationship. The query also runs very long-
SELECT fx100 AS e100048, fx101 AS e100135, SUM (fx107) AS e100620_sum,
SUM (fx102) AS e100363_sum
FROM (SELECT SUM (o100611.quantity) AS fx107, o100611.cus_key AS fx108,
o100611.day_key AS fx109
FROM ebiprod.bil_fact_billing_v o100611
GROUP BY o100611.day_key, o100611.cus_key),
(SELECT SUM (o100357.sale_qty) AS fx102, o100357.cus_key AS fx104,
o100357.day_key AS fx106
FROM ebiprod.pos_fact_retail_sales_v o100357
GROUP BY o100357.day_key, o100357.cus_key),
(SELECT o100019.chain_code AS fx100, o100107.cal_date AS fx101,
o100019.cus_key AS fx103, o100107.day_key AS fx105
FROM ebiprod.cus_dimn_customer_v o100019,
ebiprod.day_dimn_cal_hdl_v o100107
WHERE (o100107.cal_date =
TO_DATE ('20050521000000', 'YYYYMMDDHH24MISS')
AND (o100019.chain_code = 278))
WHERE ((fx103 = fx108(+) AND fx105 = fx109(+))
AND (fx103 = fx104(+) AND fx105 = fx106(+)))
GROUP BY fx101, fx100
Questions:
1) Is this an example of Fantrap ? i.e. Discoverer detects a fantrap and creates the query accordingly to resolve the Fantrap.
2) I don't think the outer join is needed and I do not understand why the conditions created in the report cannot be applied to every nested SQL. How do I change the way Discoverer generates queries ?
3) If any one has experience in enabling Drilling across in Discoverer and how it is implemented ?
Any help would be appreciated
Thanks
Manav

The joins in the SQL are controlled by the joins made in the admin edition. A check box determines if it is outer joni, one to one, one to many. The outer join would not be present unless it was present in your Business area. Click the options tab in the join in Admin.
As for a fan trap, I have always thought of it as a cyclical join. Discoverer can't determine which path to take to get the correct information. If you have the option turned on it will not allow you to build the workbook if a join such as this is determined.
The query you attached appears to be a crosstab in-line view. Try the workbook in a page detail and see if the query time improves.
One other method I have used is a hierarchy in a DW with multipath branches. This allows for drilling from summary into detail down a predetermined path. It has worked well for me but does take some effort in design to make sure the drill down is ALWAYS true, no exceptions.
HTH.

Join fact table with higher dimension level

how do i join fact tables with higher dimension levels with discoverer?
fact with detail at level C
measure X
dimension with
D->C->B->A
E->C
level
A B C
1------1------1
2------2------1
3------2------1
join between fact X and dimension level C
X=3*C because of sum(X) in discoverer and 3xC in dimension
is there a way to get correct values for X without creating a dimension like
D->C
E->

another way of asking this is whether you can create a summary table in Discoverer at a higher level than a dimension's fundamental grain. In other words - the summary examples in the documentation all describe leaving out one or more of your dimensions... they are either left in or completely taken out. But, some of the most effective summarization occurs when you summarize daily data to a monthly level. Assuming that I have a sales table (at a daily level, and a key value sales_date), and a table date_dim (primary key sales_date), I would like to create a summary sales_month_summary where the sales are grouped on month_year (which is a field in the sales_date table).
How is this done? I suspect that we can't use the date_dim table with the summary (due to the problems noted by the poster above). Do we have to create another table "month_dim"? Do we have to fold all of the desired date attributes (month, quarter, year) into the summary? Obviously we'd like to re-use all of the pertinent already existing date items (quarter, month, year, etc.), not recreate them over again, which would result in essentially two sets of items in the EUL. [One used for this month summary, and another used for the detail.]
I searched the forum - someone asked this same question back in 2000 - there was no answer provided.
The only other thought I have is to "snowflake" the date_dim into two tables and two folders, one at a date level, another at the month level. Then the detail tables can connect to date_dim (which is linked to month_dim), while the summary data can connect directly to month_dim.

So why to use Discoverer ?

Dear all,
after working as a developer since 1986 with Oracle RDBMS Rev.4 and all those Application Tools (starting with Forms 2.0, raf(rapid application facility) or whatever it was called) I wonder should I talk my customers into using Discoverer Version WHAT.
Now :
     I'm ready to go the 10g lane.
     but, what about connecting to 8.1.7
     but, look at the staus of this forum : just questions, hardly answers
     does all this web hype work right now ?
     do you have to be a super java/xml/html engineer ?, I'm just a elementary particle physicist, making his money with Oracle rdbms and Tools since almost 20 years.
     any 10-version experiences available ?
     any 9-version solid experiences ?
     I'm ready to jump, any water in the pool ?

Rainer,
Hello to a fellow old timer. Started in '88 with Oracle Add-in for Lotus 1-2-3. Then joined Oracle to support RDBMS 5.1 and SQLForms/Report/Menu and other products.
Now a Product Manager for Discoverer so I have an obvious bias so take it as it is.
Q: I wonder should I talk my customers into using Discoverer Version WHAT.
A: Yes please do and make it Discoverer 9.0.4 included in Oracle Application Server 10g (9.0.4)
Q: but, what about connecting to 8.1.7
A: no problem -- yes Discoverer will run against DB 8.1.7
When running against higher versions like 9i and 10g Disco
will take advantage of enhanced functionality like Enhanced Data Aggregation and incremental refresh of Materialized Views.
Q: but, look at the staus of this forum : just questions, hardly answers
A: Well I can give you 7 answers right now. Just trying to keep up with the high demand and limited staff.
Q: does all this web hype work right now ?
A: Absolutely. It's pretty cool actually. We have a builder/analysis tool that downloads as a Java applet (Discoverer Plus); a Viewer tool (pure HTML from a browser) that allows you to run worksheets and analyze/drill/pivot etc.; and integration with Oracle Portal so you can include Discoverer content in a Portal Page.
Q: do you have to be a super java/xml/html engineer ?,
A: You don't need to do any coding to use Discoverer.
The most challenging part will likely be designing the
Metadata (Business Areas in the Discoverer End User Layer) to enable your end users access to their data. If you want to customize Viewer you can modify an XML file if you really want to code -- but it is not required. Should be a piece of cake for a particle physicist!
Q: any 10-version experiences available ?
Q: any 9-version solid experiences ?
A: I assume you are referring to Application Server versions. We have successful customers on 9.0.2 and 10g (9.0.4) but we obviously are encouraging customers to get up to the 9.0.4 release. It has the most functionality and is very stable.
Q: I'm ready to jump, any water in the pool?
A: Come on in. The water's fine.
Discoverer Product Management

DrillDown:Is it possible using Discoverer?

Hi All,
I have a requirement in which, if there are so many results on a page, and all are links.And, clicking on a link gives related information about that link, in which it may have other links, which can again be clicked to get more information, and so on.
I came to know that it is not possible to achieve this kind of functionality using Discoverer.
Can anybody tell me that if it is possible or not and if it is possible than how to achieve this.
Plz provide me some help,if some links are available , then it will be very useful.
Thanks in advance.
Praphul.

Hi Praphul,
Yes is it possible in a couple of ways. By the way you're talking more like drill through or across than drilldown, drilldown is like year, month, day... or Suite, Product, Subproduct. Whereas you want more like Master, Detail via a link right?
- You can create a column that contains a URL (type "FILE" in EUL) that goes to another workbook/worksheet.
- You can drill to related item (standard functionality)
- You can drill to Component Rows (standard functionality)
Regards,
Gareth
Blog: http://garethroberts.blogspot.com
Web: http://www.virtuate.com

Format Dimensions in Resulting Drill Across Spreadsheet, Linked Partition

Hello,
Does anyone know if / how I can format the order of the dimensions that appear in the Source of a linked partition, when drilling across from Target to Source in the essbase excel add-in? I am in version 9.3.1, using a linked partition going from aggregate storage to aggregate storage.
An example: I have Cube "A" at the intersection where I can use my linked partition to drill across to Cube "B". When I do so, the drill works as expected. However I would like to specify an order for the dimensions of Cube "B" to show up in, in the new excel worksheet. Right now I can't determine the default order that is appearing, (it is not largest dense or largest sparse first). The order is consistently the same though, regardless of the format of the intersection of Cube A when I initiate the drill across.
I want to specify an order mostly because I want to try to automatically drill to generation 2 on a dimension in Cube B that is not in Cube A. I haven't been able to do this with the Area and Mapping defintions in the partition yet. (Maybe that is my best bet) I'm hoping there is a config file or something that I can use to manipulate how the new worksheet appears.
Can't find anything in the documentation yet.
Any thoughts would be appreciated.
Thanks,
Carmen

user616766 wrote:
Does anyone know if / how I can format the order of the dimensions that appear in the Source of a linked partition, when drilling across from Target to Source in the essbase excel add-in? I am in version 9.3.1, using a linked partition going from aggregate storage to aggregate storage.I don't believe you can control the order/etc of the dimensions in the linked partition drill-through. I just took a look at both the Java API code for the IEssLinkedPartition class and I didn't see any methods/properties that give you any information about the orientation of the members (or even any mention of the members at the datapoint).
I want to specify an order mostly because I want to try to automatically drill to generation 2 on a dimension in Cube B that is not in Cube A. I haven't been able to do this with the Area and Mapping defintions in the partition yet. (Maybe that is my best bet) I'm hoping there is a config file or something that I can use to manipulate how the new worksheet appears.The second problem you will have is automatically drilling to generation 2. The classic add-in gives you no notifications, or 'events', to tell you something has happened.
Both of these issues are trivial to solve in our Dodeca product. Our 'intelligent navigation' feature allows you to navigate from a datapoint to a new view/report and will automatically package the members represented at the datapoint as a tokens that represent the point of view. The target view/report, which you can layout with the dimensionality located whereever you wish and can connect automatically to your 'linked partition' drill-through database, then places those tokens in the location you indicate when the view/report opens. As a bonus, you do not need to specify a linked partition to have this drill-through behavior. Finally, there is robust event support so you can have the zoom-in occur automatically. You can learn more about Dodeca at our website, www.appliedolap.com.
Tim Tow
Applied OLAP, Inc

Error formula SUMCT with drill across.

Hello everyone,
I have a problem with the formula SUMCT.
A formula has been defined applying the SUMCT operator to the 0???? index 0DEB_CRE_LC.
The 0COSTELMNT characteristic is on the query rows, and a hierarchy is active on the
cost element.
When the query is executed, the result of the SUMCT operator is correctely calculated.
However, if an additional characteristic (0BUS_AREA) is drilled across, the result of SUMCT
is 0 for some hierarchy nodes.
Nevertheless, the result is correct for all hierarchy nodes when the same characteristic is drilled down.

The documentation about the [percentage functions|http://help.sap.com/saphelp_nw70/helpdata/en/e2/16f13a2f160f28e10000000a114084/frameset.htm] (which is referred by the SUMCT documentation) explains that:
If a characteristic has been drilled down in both the rows and the columns, this reference is not unique. The system displays a warning and the relevant symbol for Data Does Not Exist.
[SAP Note 164352 - Norming, Normalization / Output in percent (SUMCT, %CT)|https://service.sap.com/sap/support/notes/164352] points out that:
When you use Function 'NORMING' / 'characteristic' -> 'Normalize to' -> 'Result' / 'Calculate single value as - norming...' a warning may be displayed: "Could not carry out all normings". The same message may occur when the formula operators %CT, %GT, %RT, SUMCT, SUMGT or SUMRT are used.
Norming to 'Result' is not unique in lists which are expanded twice (both horizontally and vertically) since you could do the norming for both the column totals and the line totals.
It is possible to determine the direction in which the system calculates ('calculate along the rows axis or the columns axis') for normalization or cumulation (definable in the properties of a structure element).
However, for SUMCT operations in Business Warehouse 3.X releases, there is no alternative. Changes are planned for the upcoming BW release.
Though the suggestion works for the normalization issue, it doesn't solve the SUMCT problem, even when working with BI 7.0.
Has anyone some info about this point? Are the promised changes to be released soon?
Cheers, [Davide|https://wiki.sdn.sap.com/wiki/display/profile/Davide+Cavallari]
Edited by: Davide Cavallari on Mar 3, 2008 2:52 PM

Drill across leads to 'X' in %CT function

I am using the function %CT which gives me the share of a key figure against the subtotal.
This works beautifully, except when I try to drill across.
Drill across results into u2018Xu2019 values.
Example:
          KF 1   %CT
A        2         33%
B        4         66%
Total   6       100%
KF1 can be split in Y / N by drilling across. I would expect:
          KF 1     KF1     %CT     %CT
          Y           N          Y          N
A        1           1         25%     50%
B        3           1          75%    50%
Total   4          2         100%   100%
But instead the % show u2018Xu2019u2026.
Does anybody know what the problem could be?
Thanks,
Jan.

Hi Jan
X comes when there is not assigned value.
In your case , it is unable to calculate when KF is drilled down / accross with Y/N. Thats the reason why 'X' is coming.
Cheers
Chanda

From Drill Down to Drill Across

Hi Forum,
I'm using drill down from parent to child of the account dimesion.
I'd like also drill across dimensions.
For example, I have a report by account and I'd like drill to flow dimension. Cause i don't want to see the flow dimension for all account. The user want to see the flow details when he means that is necessary see it (unpredictable).
Is it possible? If yes how it works?
Thanks
Best regards
Gianluca

Hi,
Im not too sure, when you say drill across dimensions....
My suggestion accoridng to my undertanding is to maintain a property called flow against in acct dimension and populate it with a "Y" for account ids for which you want to display the flow dimension.
Use simple IF formula in your control sheet or anyother excel aoning with your EVDRE sheet like below and apply the same inn your flow dimension expansion.
If Flow_Property of Acct is Y
then
Flow_ALL in Expansion .
else
flow_none.
Also, If it is not based on accounts, and you wantto kae it dynamic, populate it with Flow_None always and give the user the option of selecting which accts he wants to select a flow id.
this can be done with the help of excel macro and the same excel formula as above.
hope this helps
sridhar

To visualize the report created using Discoverer Plus in Web

Hi,
I had created the report usingOracleBI Discoverer Plus Version 10.1.2.45.46c & able to visualize the report in OracleBI Discoverer Viewer Version 10.1.2.45.46c.
URL fo accessing the report is as:
http://machinename1:7777/discoverer/viewer
If i try to connect to from another machine (machine2) as
http://machinename1:7777/discoverer/viewer
i'm unable to connect & i get the error page can not be displayed.
Is it necessary to have the URL only with domain name as
http://machinename.domainname:portno/discoverer/viewer
I had already installed OWB, Oracle BI etc. If i had to create a domain then i have install all these & need to work from scratch.
Is there any way to access the report across the internet or web.
Plz do provide a solution for this issue.
Thx in Advance

Hi,
Check the Discoverer OBEs at:
http://www.oracle.com/technology/obe/obe_bi/discoverer/discoverer_1012/index.html
The lesson titled 'Creating a Business Intelligence Dashboard Using Discoverer Portlets' should answer your question.

Difference between drill down and filter using navigation attributes?

hi all,
can anyone let me know the difference between filter and drilldown options using navigational attributes. pls explain with examples for filter and drill down and how it varies.
thanxs
haritha

Jst to add up with an example :
Say you have a cost center variance analysis report with cost elements in the filter. ( not present in the default report result )
Filter - You can right click on a particular cost center & say " Keep Filter Value " - it will result in only showing that cost center & you can wish to select other filter value doing " Select Filter Value ".
Drill Down - If u wish to see the cost elements associated with cost centers you can right click on Cost Center & do " Drill Down To - Cost Element " or Drill Across.

Join fact with another fact like a dimension

Does it work if I join fact table with another fact table as a dimention(created the dummy hierarchy and add the sum() to the measure existed in the table)?
If I pick the measure value in the fact table(which is configured like a dim), I will get the correct number? Or multiplied number?
If there is any good way of work around please let me know
Thanks
Edited by: user12077461 on Jun 29, 2011 7:39 PM

Hi,
You can create an Alias Table in the Physical Layer for the Fact table which you want to use as a Dimension Table. And start using the Dimension Table. And if you want any measure values to be used from the Fact, then you can extract that too.
This should simplify your work too.
Regards
MuRam

Query across two facts with filter on unconformed dimension

Hi
This question is actually regurgitated from another post as I wrongly tagged it onto the end of an already answered question.
Anyway, I am trying to query across 2 fact tables which have a dimension in common (conformed) and also some dimensions that are not directly joined to BOTH fact tables (seemedly referred to as 'unconformed' in OBIEE).
So in it's simplest form
Fact1, measure1 linked to dim1 & dim2
Fact2, measure2 linked to dim2 only
I have followed the recommended method advertised here and elsewhere of adding the second fact as a LTS of the first in the BMM layer and setting logical levels on measure2 to Total for all dimensions that fact2 is not joined to (in this case dim1).
When a basic query is issued in Answers that filters on the unconformed dimension (dim1), a spurious result set is returned that includes all records for the second fact.
This is quite a standard warehousing reporting query whereby 2 sets of data are queried independently and then joined on their common attributes but in this scenario the BI server seems to always generate a FULL OUTER JOIN and return an unrestriced data set.
Has anyone else come across this strange behaviour?
thanks
Tony

adding the second fact as a LTS of the first in the BMM layer
Could you try without this step? Just have separate facts and separate LTSs and set
Fact1 LTS to lowest for Dim1
Fact1 LTS to lowest for Dim2
Fact2 LTS to Total for Dim1
Fact2 LTS to lowest for Dim2
?

Drill Across fields mapping with calculate field

I create a calculation field in one view, then when config drill across, i set the calculation field to mapping to a Parameter of target view
Save report
Then, i open the report in view mode, click the view which config drill across, it get JavaScript error in DrillAcrossModifierManager.js line 272, Object required

Hi,
I'm using BAM 11.1.1.4, and I have a report wich calls another report through DrillAcross.
Both reports have no calculated fields and I am receiving the same error message:
Mensagem: object required
Linha: 273
Caractere: 5
Código: 0
URI: http://server:9001/OracleBAM/13846/reportserver/scripts/drilling/drillacross/DrillAcrossModifierManager.js
Any idea about the cause of this error?
Thanks.
Luciano Gomes

How to drill across sections in Dashboards

I hav got this through this site.
(if i click on a department name in department section in dashboard, the corresponding employees details should be shown in the adjacent column/section without rendering to a new page)
http://oraclebizint.wordpress.com/2007/12/26/oracle-bi-ee-101332-drills-across-sections-in-a-dashboard/
i hav got the total thing done.
i'm doing this for an department table on department name.
for which when i drill on department name, corresponding employees list should be shown.
my problem is
1. if the dept name contains a spaces like 'Human Resources', then the URL is taking only Human and its ignoring the rest.(if it doesn't contain spaces, its showing correctly.!)
the URL looks like this
http://l3f0372:9704/analytics/saw.dll?GO&path=/shared/New/3.Emp&Action=Navigate&P0=1&P1=eq&P2=Departments.%22Department%20Name%22&P3=Human
2. the drill down can be implemented on varchar datatypes only.
but not on number types like dept_id. Is there any way to that?
Edited by: user10564344 on Nov 23, 2008 10:48 PM

Thank you Naresh for responding to my problem..
But, if we use Navigate, the drilling will be to a new page.
but the client requirement is like this.
intially there are two sections, one shows dept details and other shows emp details(of all depts)
when we click on a particular dept name, the emp details of that department only should be shown and in the same section only(i.e., without navigating to a diff page).
the procedure i'd got from the website mentioned in the earlier post worked fine with the department name containing no spaces.
but when there is a space in the department name(ex.Human Resources),, the URL is taking till Human only(due to anchor tag in column properties on dept name).here the anchor tag is used to call emp details section and pass the respective dept name called.But when a space occurs in URL, that anchor tag is taking till a space.
(ex.: http://l3f0372:9704/analytics/saw.dll?GO&path=/shared/New/3.Emp&Action=Navigate&P0=1&P1=eq&P2=Departments.%22Department%20Name%22&P3=Human )
so, how to pass a value with space using anchor tag dynamically..?
Thanks in advance..:)

Drill across (joining facts) using Discoverer

Similar Messages

Maybe you are looking for