Performance - Conversion of characteristic values to SID

Greetings:
Yesterday I load 556.000 record into an infocube.
No transformations of any kind. Direct assign for every infoobject. Very simple
It lasted 1 h 30m to complete the loading.
It started loading 7 datapackage of 50.000 records each.
I realized that 3 of those datapackage (#2, #4 and #7) lasted from 41:47min to 41:59min in the "Conversion of characteristic values to SID" task. The other 5 lasted ... seconds.
Any idea of that could be going on during this conversion?
I don't understant why some datapackages last so long and other don't.
Does anybody has any idea about this?
Thanks in regards.
Guillermo

Hi Daniel,
The SID generation is based upon the Dimension Table and Master data.
If you see the dimension table can be build with different types like:
one to many
many to one
many to many
one to one
So, if the dimension's for that datapacket data is many to many then it has to get all the relations and then it has to assign an SID.
If it is one to one then there will be no more work to be done by the system to assign the SID's
Hope you understood the back-end process.
With Regards,
Ravi Kanth.

Similar Messages

Conversion of Characteristic Values to SIDs is taking long time

Hi Experts,
Cube load is taking more than 7 hrs to complete for 6 million records. Max time is at Conversion of Characteristic Values to SIDs. Cube consists of 500 fields. Load is from DSO to Cube. DSO has 10 fileds with no SID flag check for activation. Remaining fields in the CUBE are pouplated looking in to other DSO's. Start and End routines are executing very fast. How to fasten Conversion of Characteristic Values to SIDs step. I need to load 15 million records daily as part of delta. any suggestions.
Thanks
Sunil

Hi Sunil,
Go through the below link, it will be useful
http://aq33.com/business-warehouse/Articles-05269.html
Regards,
Marasa.

Slow dataload into Infocube- Conversion of Characteristic values to SIDs

Hi All,
we are facing a slow delta load issue when loading from a DSO into an InfoCube ( approx 150,000 recs).
After checking the monitor found that
- The Conversion of Characteristic values to SIDs is taking time
- The Start Routine time is real quick ( in secs)
The underlying DSO is hugh and there is no SIDs Generation upon Activation flag set for it
Master data is loaded first and than we are loading the transaction data.
i've tried refreshing the stats of the Cube, Reduce the size of the data package,
Deletion of Indexes than Loading and recreation of Indexes, no joy at all.
Any comments and recommendations are much appreciated.
Thanks,
Natalie

Hi Natalie,
The problem for your slow delta load from DSO to cube is because the SIDs Generation upon Activation flag is unchecked at DSO level .
Becasue of which your activation is faster but your loading from DSO to cube will be slower as you only mentioned that SIDs are being generated .
One recommendation will be to switch on the property of the DSO SIDs Generation upon Activation flag .
Are there any secondary indexes also in DSO ?
You can try deleting them also during loading the data from DSO to cube .
I guess your change log table is also huge in this case . If possible we can think of deleting the log table also . Then for the new records let it get filled .
Hope the above reply was helpful.
Regards,
Ashutosh Singh .
Edited by: DEADLOCKS on Apr 21, 2011 5:08 PM

Conversion of Characteristic Values to SIDs. Load Very slow

Hi Experts,
We have a slow load issue when loading from a Write-Optimized DSO into an InfoCube. 8 millon recs in 10 hours.
We check the start, end, and rules routines and every is ok but for all the package the load was very slow in the Conversion of Characteristic Values to SIDs.
Can any help me with this?
Thanks a lot,
Alejandro

Hi,
It is because the SIDs Generation upon Activation flag is unchecked at DSO level .
One recommendation will be to switch on the property of the DSO SIDs Generation upon Activation flag .
Are there any secondary indexes also in DSO ?
You can try deleting them also during loading the data from DSO to cube .
Also load your master data first and activate it before loading it to cube.
Thanks,

Conversion of Characteristic Values to SIDs

Does anyone know why this DTP step would take 15 - 20 min for 10,000 records?

Hi Hank,
Just check out if there are any Indexes created on that cube,if yes pls delete the indexes and Load the Data and then Create the Indexes.
Also check with the Basis team for the Server Load,which shud not be a issue as u said this is the Only job that is Running.Hope u have loaded master data before transaction Data.
Rgds
SVU123
Edited by: svu123 on Jan 4, 2010 11:42 AM

Conversion of characteristic values before displaying in Bex Analyzer

Dear all,
does anybody know whether it's possible to get influence on the manner in which a characteristic is embedded into an BEx Analyzer sheet (may be via user exit, function module exits or something else) .
In our project we have characteristics (describing production parameters) with numeric values which can't be used successfully with EXCEL AutoFilter functions because they are strings in EXCEL. Due to our data warehaouse design koncept it is not all right for us to store these values as key figures. So we are searching for methods to convert these string values into numeric values before we give them into the EXCEL sheet.
Lot of thanks for you in advance!
Dorothea Gebauer.

High Leo Gillian,
at first thank you very much for your answer. Unfortunately your information is not practicable for us. There are hundreds of characteristics, in the future there may be mor than two thousend, which have to be converted before they are embedded into the sheets. Therefore we need an automatically done conversion.
with best regards Dorothea Gebauer

When we run query how it gets the characteristic values and attributes.

Hi,
When we load transaction data it chacks characteristic values then SIDs then DIM IDs then insert DIM IDs into fact table but when we run the query how it checks and gets the characteristic values and attributes.
Bye
GK

when we run the query how it checks and gets the characteristic values and attributes.
Just the opposite way you have described. It gets from the corresponding masterdata tables, with the connected SID.

RSRV Check Check if characteristic values of text table exist in SID table

Hello Guys,
I have run the RSRV check Check if characteristic values of text table exist in SID table and I got a warning
when I check the Text table and compare it to attribute table, total entries in text table is more than entries in attribute table..
TEXT TABLE DONT HAVE ANY LANGUAGE...
is it possible that there are more entries in text rather than attributes... again THERE are no language..
Many Thanks

try to do a 'select * from table' and catch the
exception as it might fail if the table doesn't exist.Which might take quite a time if the table is big! If you really want to do that I'd use SELECT * FROM table WHERE 1=2
A better way is, to ask the DatabaseMetadata object to retrieve information about the table...
Thomas

Characteristic value '5000000000 ' of characteristic 0MATERIAL is not MATN1

Dear Experts,
I am getting one error message while extraction,when executing the DTP .the messge is :
"Characteristic value '5000000000 ' of characteristic 0MATERIAL is not MATN1 -converted" BRAIN 68 @3R@
what is the meaning of this messege ,Request you to plese suggest me reagarding the same .thanks in advance .

Hi Bhullar,
Have you maintained material length as 18 at ECC and BW sides.
at bw side info object 0material conversion routine as MATN1. seems like you may be maintained as ALPHA. if your using as ALPHA conversion routine then at transfer rules for MATERIAL you can pass output data by suing FM - CONVERSION_EXIT_ALPHA_INPUT.
or if possible then change conversion routine to MATN1 for material(info obj) at data source level or info object maintenance(RSD1) level.
Thanks

Condition type value updation using characteristic value in sales order

Hi Experts,
We are implementing SAP for one of our Spain client. In that one of the Legal requirements is, whenever they are selling the finished products to domestic customers who contain OIL as one of the component then, some additional cost needs to be calculated and collected from customer and which needs to be paid back to Spain Government every month on Region basis.
If OIL is not a component to produce the finished product then, not required to charge the customer additionally.
The OIL content used to produce every finished product will vary based on product characteristic value. Meaning, for some finished product the OIL used will be 2.2 KG and for some products it will be 3.5, 4.3 KG and etc.
Since various combinations of products are available for the client, during sales order creation we are planning to update some special characteristic which the data flow from u201CVariant Tableu201D. In the variant table we are going to maintain the OIL Quantity required making finished product.
But we are not sure how to activate a new condition type with the value fetched in the special characteristic. This new condition type needs to be activated based on this characteristic value and multiplied with total order quantity and price to charge from customer.
Example, the special characteristic fetched the value of 2.2 KG based on variant table. Total Order quantity is 10 PC and charge is 0.06 EURO per KG. in this case, (2.2 x 10 ) x 0.06 EURO = 1.32 EURO
I need the expert opinion how to activate the new condition type based on characteristic value and achieve the above requirement.
Thanks in advance.
Warm Regards,
VEL

Hello,
I have replicated the whole requirememt in IDES and getting the price calculation as needed but the only issue with the Conversion factor for the condition type. Either you need to maintian the conv factor in material master or need to enter manually for the condition type in SO. Some how you have to maintain the conversion factor to calculate the price for the condition type in different unit.
For easy of understanding Lets make the whole requirement in 2 parts
Part - 1
Oil qty calculation. For convenience sake I have taken oil qty also in TO
1) Created a num char OIL_QTY1 with length 9 and decimal places 5 and UOM as TO
2) Created Variant table to infer the OIL qty in TO per piece of finished prod ( This oil qty data may be depends on various factors). For certain Finished products the oil qty may be 0 or blank
3) Created a REF char SO_QTY to pull the SO qty to configuration and add this char in class 300(Table:VBAP,Field :KWMENG)
4) Created another char SO_OIL_QTY (Num with length 9 and decimal places 5 and UOM as TO)
5) Created a PROCEDURE say PROC_SO_OIL_QTY with code
*$self.SO_OIL_QTY = $self.OIL_QTY1 * $self.SO_QTY*
6) Create another PROCEDURE as mentioned earlier with modified code
$SELF.A_VARCOND = 'TEST_001' if OIL_QTY1 specified,
$SET_PRICING_FACTOR ($SELF,A_VARCOND,'TEST_001', SO_OIL_QTY)
Note: Maintain proper seq of the procedures in profile. The Qty cal proc should be before the price related procedures.
PART-2
1) For the condition type maintained 60 EUR per TO (0.06 EUR per KG)
Now if you create SO for the Finished product of 10 PC
the Char values will be
OIL_QTY1 = 0.00022 TO (inferred from Variant table th proc)
SO_QTY = 10
SO_OIL_QTY = 0.0022 TO
Get into conditions tab
Here Condition type ZSIG will be displayed as error with Amount 60 EUR per TO and condition value will be blank. Just enter value 1 in the col NumC and CCon fields.
Now the Condition value will appear as 1.32 EUR.
Hope this gives some more clarity
Regards
Brahmaji D

All Dates displayed for "Characteristic Values" in Query Designer

Hello Experts,
We are on BI 7.0, level 13. I am having an issue within the Query Designer with regards to dates. I have a write-optimized DSO that contains 3 date fields (for example, ZDATE1, ZDATE2, and ZDATE3). Each date InfoObject is of type DATS so the system automatically creates it with reference to 0DATE.
When I create a query in the Query Designer, on the left hand side, I expand the "Characteristic Values" node under each date field. The Query Designer shows the same list of values for each of the 3 dates even though they are not valid values posted in the DSO for those fields.
For example, ZDATE1 only has 1 value posted in the DSO (01/01/2005).
ZDATE2 only has 1 value posted in the DSO (01/01/2006).
ZDATE3 only has 1 value posted in the DSO (01/01/2007).
Bute when I expand the "Characteristic Values" node in the Query Designer, I see ALL THREE values under each date field. I would expect to only see the 1 value posted for the InfoObject in the DSO. Also note that each InfoObject is defined to show "Only posted values in InfoProvider".
It appears that Query Designer will show all values for the reference InfoObject 0DATE instead of the ones posted to the actual InfoObject in the DSO. If I delete the data in the DSO, the Characteristic Values list still remains because they exist in 0DATE. Anyone encounter this before? How can I get the Characteristic Values list to only show posted values for that InfObject?
Thanks for your help!
J

Thanks for the response. I went into the DSO and right clicked on each of the Date fields. I looked at the Provider-specific properties and there is the option for "Query Exec.FilterVal" which only restricts what values appear when restricting during execution, not during query creation.
Is there someplace else I should look or someplace else I can make change to only display posted dates when creating a query? Thanks!

Mapping conversion of source value 4448601.75 to target value 448601

Mapping conversion of source value 4448601.75 to target value 4448601 or source value 999999.99 to target value 999999
Here standard functions like round, floor, ceil..etc, will not work because they will convert source value 4448601.75 to 4448602 and source value 999999.99 to 1000000.
Also i may have blank or white spaces coming in front of the source value from R/3 side like _____999999.99 and i want the value at the target side as _____999999. Here _____ representing blank or white spaces.
How i will do the mapping, can someone help me out.
Thanks,
Ashish
Edited by: Ashish Soni on Sep 6, 2008 9:53 PM

>>Here standard functions like round, floor, ceil..etc, will not work because they will convert source value 4448601.75 to 4448602 and source value 999999.99 to 1000000.
No....! The floor function can be used. It returns the largest integer that is not greater than the passed value. However this will also trim the spaces.
So u can write a UDF that accept the number and substring till the decimal point and returns it.
Try this.. It works..
public String converv(String a,Container container)
return a.substring(0,a.indexOf('.'));
Thanks
SaNv...

Numc characteristic values are not saving in AUSP

Hi All,
I am unable to save numc/Date type characteristic values in AUSP table using CL20N tcode. i am able to retrive CHAR values from AUSP table but NUMC / DATE values are not saving in this table.
Regards,
Praveen

Hi Praveen,
If you are fetching ATINN from AUSP table, then you need to use conversion routine..
Call the FM CONVERSION_EXIT_ATINN_* & you will get the converted value.
Thanks,
Sharath

How to sort the characteristic values in layout

Hello,
I wont to use a layout with key figs in data columns and two characteristics in lead columns. The problem is that the sorting of first characteristic values is displayed in alphabetical order. How can I change the order to another exact defined combinations for characteristic values?
I read the thread with the same problem (see below the link), but the solutions are not satisfacroty.
Has anyone another idee?
Thanks for any help!
Larissa Bräutigam

Hi Larissa;
I had the same problem and they propose this:
Attributesu0092 order that the system displays in the Planning Layout
That was not the solution presented to the customer. Because of system performance, a programmer was there and created a VBA code. But I wasn't there so I don't have the code.
regards

Extended/modified to accept characteristic values for BAPI_SALESORDER_CREAT

Hi All,
can any help me extended/modified to accept characteristic values for BAPI_SALESORDER_CREATFROMDAT2 BAPI.my requirement is
Exploring the standard BAPI BAPI_SALESORDER_CREATEFROMDAT2 (or any other suitable)to be extended/modified to accept characteristic values
and based on characteristic values, it should perform variant matching
and if in case variant matching is successful, the BAPI should replace the KMAT material with the exact matched variant material.In case the variant matching is unsuccessful, then retain the KMAT material in the sales order line item

Hello Nagarajan,
I have used this BAPI in my last task. You are passing the values correctly. I will tell you why it is giving like that message.
Whenever we will create a sales order in VA01 transaction, you will get a message like "The sales order is not complete, would you like to EDIT" while saving the sales order.... This is because of not providing complete data required for careation of sales order.
Try to implement the changes manually in VA02, if you get same errors, then you can find your code is correct.
Even I got same type of messages when changing a sales order to change the Rejection Reason for the items.
Best Regards,
Sasidhar Reddy Matli.
Edited by: Sasidhar Reddy Matli on Jul 31, 2008 2:20 PM

Performance - Conversion of characteristic values to SID

Similar Messages

Maybe you are looking for