Exception Aggregation - Performance
Guys,
We have a query with ten calculated key figures and all of these have exception aggregation of summation at material level. Currently for just one plant the query is taking 8 minutes of time which is too bad.
These 10 CKF's caluculate different types of stock values..all these have summation as exception aggregation. The price is calculated on the fly by using replacement path attribute value. Any suggestions on how to increase the performance of the query or any other approach to handle this scenario.
stock value 1 = quantiy1 * price1.
stock value 10 = quantiy10 * price10.
Please let me know if you need more info.
Thanks,
Varma.
Hi Varma,
Make sure that you have these Notes in the system:
1416737
1387593
1396485
Regards,
Michael
Similar Messages
-
Query Performance with Exception aggregation
Hello,
My Query Keyfigures has exception aggregation on order line level as per requirement.
Currently cube holds 5M of records, when we run query its running more than 30min.
We cont remove exception aggregation.
Cube is alredy modeled correctly and we dont want to use the cache.
Does anybody can please advice if there is any other better approach to improve query performance with exception agg?
ThanksHi,
We have the same problem and raised an OSS ticket. They replied us with the note 1257455 which offers all ways of improving performance in such cases. I guess there s nothing else to do, but to precalculate this exception aggregated formula in data model via transformations or ABAP.
By the way, cache can not help you in this case since exc. agg. is calculated after cache retrieval.
Hope this helps,
Sunil -
How to find the list of Queries/Reports which are using Exceptional Aggregation in SAP BI?
Hi All,
We are interested to know how to find the list of Queries/ Reports which are using Exceptional aggregation in SAP BI.
Please let us know is there any table's to check the list of reports using Exceptional Aggregation in SAP BI.Hi,
Here you go..
1) Go to table RSZCALC and get list of ELTUID where AGGREXC is not INITIAL and AGGRCHA is not initial; now you get only exception aggregation set based on some chars. Also you can further add STEPNR = 1 since your intention is just to get query name , not the calculation details; this will reduce number of entries to lookup and save DB time for next steps.
Here you will get list of exception aggregation UUID numbers from which you can get properties from RSZELTDIR.
2) Pass list of RSZCALC-ELTUID to table RSZELTXREF - TELTUID and get list of RSZELTXREF -SELTUID - this table stores query to it's elements maping kind.
3) Now again pass RSZELTXREF - SELTUID into same table but into different field RSZELTZREF - TELTUID and get RSZELTXREF - SELTUID
This step you get query reference sheet or column or query general UUID for next step.
4) Pass list of RSZELTXREF - SELTUID into RSZELTDIR - ELTUID with DEFTP as 'REP'. Now you get list of query names in RSZELTDIR - MAPNAME and description in TXTLG.
Note: you can also get the reference chars used for exception aggregation from RSZCALC - AGGRCHA field.
Hope this helps.
Please keep in mind, it might take more time depends on how many query elements you have in the system...
Comments added for better DB performance by: Arun Thangaraj -
Query Problem With Exception Aggregation
Hello,
I have two I/O 0Customer and 0Cont_Acct and also I have a Infoset made out of both. Here is the problem i am trying to solve,
I want to figure out the customers who has more than five contract accounts associated with each customer number.
Here is the solution I approached....
1) Created a formula variable counter on the KF # of data records (which is provided in the Infoset automatically ) with exception aggregation total while referencing the contract acct I/O.
2) Kept my customer # and Contract acct on my rows area.
The results I am getting is less than satisfying. I am expecting this output below.
Customer # Cont Acct Total No of Cont.acct
A 1001 1
1002 1
1003 1
3
B 1007 1
I would appreciate anybody's help in this. Thanks a lot.
-AHi,
We have the same problem and raised an OSS ticket. They replied us with the note 1257455 which offers all ways of improving performance in such cases. I guess there s nothing else to do, but to precalculate this exception aggregated formula in data model via transformations or ABAP.
By the way, cache can not help you in this case since exc. agg. is calculated after cache retrieval.
Hope this helps,
Sunil -
Exceptional Aggregation in Query Designer
Hello Gurus,
we have an existing Report at the Material Level on the InfoCube.
Each Material may have multiple POs and we are having multiple
records for each PO varying the date. Now the user want to view
only the record latest changed Record at the PO level
and Sum of all the records at the Material Level.
For this i have created a CKF using the Exception Aggregation
as Maximum for the 0calday. But client want the Report at Material Level
he dont want the PO's details. When i remove the PO from the report
it gives the result for only the latest PO not all the PO for that Material.
For Example we are having below records in the cube.
PO Material Calday Qty Price
101 ABC 1/12 100 10
101 ABC 2/12 100 10
101 ABC 3/12 100 11
102 ABC 4/12 100 10
104 ABC 5/12 100 10
103 ABC 7/12 100 11
I am getting the result with PO in Rows as Below
PO Material Calday Qty Price
101 ABC 3/12 100 11
102 ABC 4/12 100 10
104 ABC 5/12 100 10
103 ABC 7/12 100 11
When I remove the PO's From the Rows then it will display
as below i am placing the calday for under standing the
question actually it is not present in the output of the below report only.
Material Calday Qty Price
ABC 7/12 100 11
Which is not currect I have to get the Result as
Material Qty Price
ABC 500 42
My Concerns:
!. Kindly let me know if it is possible and what are the
other things that i need to consider.
2. Did it will impact the performance.
Waiting for your quick replies
Thanks & Regards
KK
Edited by: KK on Dec 23, 2009 11:08 AM
Edited by: KK on Dec 23, 2009 11:15 AM
Edited by: KK on Dec 23, 2009 11:22 AMHi KK
Confirmation:
First I need to confirm one thing,
as you mentioned expected output is,
Material Qty Price
ABC 500 42
But I think, it should be
Material Qty Price
ABC 400 42
right?
Answer:
Can you please try below steps and let me know whether it is working as expected or not
Create a basic Key figure for calday and include it in your InfoCube. In the transformation, assign it to same fileld which is a source of calday.
In query, Create a calculated key figure based on calday key figure with exception aggr as last value and reference characteristics as PO.
Regards
Rohit -
How can i do exception aggregation please give scrreenshots steps in bi-7
how can i do exception aggregation please give scrreenshots steps in bi-7
HI Deba,
We can create exception aggr in inof object level........
we can maintain at query level also.........
Exception aggregation
In the query definition, the exception aggregation that was specified with the InfoObject
maintenance can be overridden. This overriding leads to a worsening of performance
however.
https://www.sdn.sap.com/irj/sdn/go/portal/prtroot/docs/library/uuid/bda556e9-0c01-0010-83b0-d519d6deb9e9
Regards,
Vijay. -
Exception Aggregation takes a lot of time
Hello, guys!
I'm having problems with exception aggregation for some of my key figures in BEX Query Designer.
Well, they are working properly (Summation and Counting), but it takes up to ten minutes for them to calculate, and this is not acceptable for both me and my client.
Without those Exception Aggregations, query runs in a seconds.
Do you know any way to avoid those problems? Maybe it is possible to do something on BW side?
I'd appreciate any advise on this.
Thank you in advance for your help,
Nik.Nikita,Exception aggregation definitely takes a toll on the performance.There might be a situation where it is aggregation huge number of records with the unique combination of reference characteristic.
You can either try to filter out the data as much as possible after analyzing the data in cube.
Also i can suggest to do a close health check up of your cube with the help of program
SAP_INFOCUBE_DESIGNS
Regards,
AL -
DIff in Exception aggregation bet 3.x & 7.x
Do we have any difference between 3.x & 7.x exception aggregation at Query level!
That is one big advantage of 7.x query. Exception aggregation is much more flexible and useful. You can use it perform calculations at detail level and display results at a summarized level. Check out the help section. It has good explanation of the 7.x functionality
-
FBL5 FIAR - exception aggregation - count customers with open items at key
Hello gurus,
I have read the information about counting all values in reference to a characteristic. I manage to count how many customers have a FIAR open item 'billing status' = Open in my FIAR cube.
Now I am asked to replicate transaction FBL5 of R3. This transaction will give you open items at a given key date by adding
the items with
A) Item status = 'O' and posting date <= key date
and the items with
B) Item status = 'C' and posting date < =key date and clearing date > key date.
I calculate the net value by creating 2 restricted key figures and then adding them in a CKF.
But for counting I have a problem. I can count how many customer make up the open items (case A)
and how many customers makeup items that where cleared between key date and today.(case B).
But I cannot sum the two numbers as there might be customers that are counted in both cases.
For example I have 4 customers that correspond to case A : 1000 1011, 1012 and 1014
And I have 3 customers that correspond to case B 1000 , 1011 and 1013
I would like to get 5 as the number of customers that verify either case A or case B (or both).
Is there a way to do this with the exception 'aggregation 'and 'count all values' property of CKF ?
ClaudioHi,
The logic given in the above post may be a bit confusing to you. Check out this explanation and try to understand what I mean.
Posting Date - PD
Key Date - KD
Clearing Date - CD
PD--KD--
CD
-O(Status)-C--O-----C
-5107-----8
Let me explain the above table.
Between PD & KD there are 5 count with status'O' and 10 counts with status 'C'.
Between KD & CD there are c counts with status 'O' & 8 counts with status 'C'.
This means 5 with status 'C' becomes 7 between KD & CD.
Similarly, 10 status with 'O' becomes 8.
The logic I am trying to say is..
If you don't want to get counts with status 'C'(repeated) as I understood) 7-5 = 2. Show this 2 as count for status 'C' between KD & CD.
So the table will become like this now.
PD--KD--
CD
-O(Status)-C--O-----C
-5102(7)---2(8).
In braces shown are the total with status 'O'/'C' for the dates.
I hope this will solve your problem.
Message was edited by: AVR - IGA -
How to find the value of date char used as exception aggregation reference
Hi BI Gurus,
On a BEx report I need to list three things by material:
1) the open (not yet delivered) Purchase Order quantities
2) the PO quantity to be delivered next and
3) the date that belongs to the next delivery
The model supports these data i.e. for each material I can list all open quantities by Purchase Order / Item / Schedule line and the scheduled delivery dates are also available as a characteristic.
Determining 1) is easy as the drilldown is fixed (materials only) the open quantities get summarized for all PO-s belonging to the materials.
To determine 2) I used a Calculated KF simply including KF 1) =Open PO Qty in the definition and setting Exception Aggregation (first value) with a reference char of the delivery date. This gives back the Open PO Quantity to be delivered first.
My question is about how to determine the 3rd value (actually this is not a KPI but a characteristic value). In other words, for each materials I would need to determine the first among all of the possible delivery date char values of the open PO Items / schedule lines. This is the date the quantity shown in the 2nd KPI will be delivered on.
Does anybody have an idea how to approach this issue?
Thanks for you help in advance,
AttilaHi Olivier,
Thanks for your suggestion. I got a bit closer with the replacement path formula variable and the CKF. I tried to apply the same logic of KPI 2) but for some reason it did not work. Actually, it delivers the right value but only when 0SCL_DELDAT (Scheduled delivery date) is in the drilldown But what I need is the first delivery date by Material only, and without this 0SCL_DELDAT drilldown.
Any other ideas?
Thanks and bye,
Attila -
Exception Aggregation help in Bex !
Hi Experts
Iam creating a query with date calculations. I have the following calculations in my qry...
1 . Cycle Time : ( Doc Date - Quote sent date) ....I created two replacement path variables and use them in formula.
2 . Nr of Quotes : ( Count of all sales orders with Block L1 & L2) ...I created a CKF for counting the nr of docs then used that CKF in an RKF to filter by block.
3 . Avg Tender Time : ( Nr of Quotes / Doc date - Quote rec Dt) .........Created a formula same as 1 with Quote rec dt instead of Quote sent then created another formula to make 2/3.
Now when I execute the query Iam getting numbers but
Cycle Time I took Exception Aggregation as Counter for all detailed values and ref char as Doc Date....Its giving some nr as output...but my question is if I take doc date as Ref char how it gets the calculation from replacement
Does any one have worked on date calculations as the above situation / any one have any how to's which explaines how to find date calculations with exception aggregations?
Thanks in advance...fixed
-
Bex Exception Aggregation in 3.5 equivalent to 'TOTAL'
Hi All,
Can anyone help me in getting a functionality equivalent to exception aggregation 'TOTAL' in BW 3.5.
I have a requirement which can only be fixed using the exception aggregation 'TOTAL', my bad in my current client they dont have BI 7.0. So can anyone tell me what is the equivalent function of 'TOTAL' in BW 3.5.
Regards,
AnandI need to develop a report which will show Org unit & degree (education data) wise head count and annual salary.
As the education records for an employee will be more than one, the annual salary will be doubled or tripled in the DSO or cube.
If I do an average of annual salary based on employee the overall result goes wrong.
Say for example, empl 20045789 as three record with KF value 10,000, the result in the report I see as 30,000.
SAY in the backend
EMPLOYEE | Degree | Annual Salary
20045788 | A | 50,000
20045788 | B | 50,000
20045789 | A |10,000
20045789 | B |10,000
20045789 | C |10,000
In the query I get it as
Org Unit | Employee | Annual Salary | Average(Annual salary/ Number of employee)
100001 | 20045788 | 100,000 | 50,000
100001 | 20045789 | 30,000 | 10,000
overall result 130,000 | 26,000 (130,000/5)
The expected result is 50,000 + 10,000 = 60,000 / 2 = 30,000.
Pls provide me solution.
Regards,
Anand
Edited by: araj123 on Jun 3, 2011 3:45 AM
Edited by: araj123 on Jun 3, 2011 3:46 AM
Edited by: araj123 on Jun 3, 2011 3:47 AM
Edited by: araj123 on Jun 3, 2011 3:48 AM
Edited by: araj123 on Jun 3, 2011 3:50 AM
Edited by: araj123 on Jun 3, 2011 2:25 PM -
How to determine the "Last Value" in exception aggregation
Hi gurus,
I have a KeyFigure with an exception aggregation -Last value- by time.
I report on a multiprovider ("containing" a infocube and a realtime infocube with BPS-InputLayout). I can correct separate Items with this BPS-Layout. So, at the end I have 2 values for one item.
So, by these facts, my report shows the the last value for the specific KeyFigure, but I expected that both values (the value in the infocube and the correction via BPS)are sumed up.
As this is due to the exception aggregation not the case, I could "overwrite" the normal value in the infocube with the BPS.
But SAP determines the normal value in the Infocube as "Last Value", not the one in the correction cube, which is entered via BPS.
How does SAP determine the "last value" as I can´t see any date...
Thanks in advance!Thank you for your answer...
I got the soltion by myself.
I thought, Last Value means the last Value, depending on the time, but in my opinion the function last value should be called "Greatest Value".
All figures with the same value for the reference characteristc are aggregated anyway. -
Problem with exception aggregation
Hi forums,
Can any one help in this issue?
We checked the data in the ECOM cube for the order 63349312 where it is giving 6 records eventhough we are deleting the adjacent duplicates of the order number in the start routine. (Is it like the deletion of adjacent duplicates happen in data packet level instead of happening in the entire cube level? B'cos the same order might present in the different packets.)
Hence the data is aggregating for the key figure, 'Number of Days' in the ECOM report. But if we check the keyfigure properties in the RSA1, there we selected the Exception Aggregation as 'First Value' and aggregation reference char as 'ZN_CORD' (Order number). Still it is aggregating the number of days value in the report.
Regards,
Charan.Hi Rohit,
We are doing same procedure as you mentioned. We assigned order number as refer characteristic to keyfigure No. of days in infi object level. But this operation is not working in report level
Order number ZCUDATE ZN_GSTRP ZN_CORDKF ZN_DAYSKF
63349312 18.01.2009 01.10.2008 1,000 109
63349312 18.01.2009 01.10.2008 1,000 109
63349312 18.01.2009 01.10.2008 1,000 109
63349312 18.01.2009 01.10.2008 1,000 109
63349312 18.01.2009 01.10.2008 1,000 109
63349312 18.01.2009 01.10.2008 1,000 109
I want o/p as
Order number ZCUDATE ZN_GSTRP ZN_CORDKF ZN_DAYSKF
63349312 18.01.2009 01.10.2008 1,000 109
but it is showing as
Order number ZCUDATE ZN_GSTRP ZN_CORDKF ZN_DAYSKF
63349312 18.01.2009 01.10.2008 1,000 676 -
Exception aggregation on Non-cumulative key figure
Dear experts,
I have a problem. We have an InfoCube in BW in which the Activity Journals are loaded every day. In this Activity Journals it is registered that a Material was listed (available) at a customer (value 0 or 1).
We have a non-cumulative key figure MXW_LISTN with inflow (MXW_LISTI) and outflow (MXW_LISTO). In time this key figure receives +1 or -1 for a customer, material combination.
I want to use MXW_LISTN in the query in a specific way. The key figure on itselve tells me how many listings there are. However, I want to know how many business partners there are with a listing.
I was thinking about an exception aggregation on MXW_LISTN with reference characteristic 0BPARTNER and "Count <> 0". BW however refuses to do so and only lets me do simple aggregations such as AVG, MIN, MAX etc.
Can you think with me and come with an proper solution to count the number of business partners with a listing?
Your help is very much appreciated and rewarded!
Best regards, Hans de KleinMaybe a solution with virtual key figures is an option?
Please help me out!
Best regards, Hans de Klein
Maybe you are looking for
-
Win 7 (preloaded) and Win XP in dual boot. Who has it?
Hi there! I've got a G550 with Win 7 (factory installed) and I need to install Win XP as second S.O. I tried some things but it's not working properly.... The disk is divided in (approx): 200 MB hidden partition (from factory) 160GB with Win 7 instal
-
Cannot turn on "Show the day of the week" option in Date & Time
OS X Lion turns "Show the day of the week" option off each time I open that preference pane and after every reboot. My region in Formats tab of Language & Text preference pane is set to "Russia (Russian)". I've noticed that the problem doesn't appear
-
Is there a way to delete multiple duplicate photos easily?
Is there a way to remove duplicate photos. I have a hard disk crask and the backup put multipul copies of photos in the photo file. Is there an easy way to eliminate them?
-
Notify Operator Task not sending email, Database mail works fine, Added Profile to Agent
SQL Server 2008 R2, Latest Updates. I added a Service Broker that has Alerts and Notifications in it. In adding the service broker I had to go into the Agent Properties, Alert System, enable mail Profile, enabled the failsafe operator, etc. I get
-
Audit fixes - username equal password
Background: Oracle 11g on Windows environment I am working on a query that will provide a list all usernames in my system where the username and password are the same. The problem is the passwords are not plan text so the query give zero result. Is t