How to initialize Datamining Model in process chain ?

Hello all,
I need to launch an APD for datamining analysis but i need to initialize the model before each data loading.
The only way to initialize the model is to write an abap program ?
No standard process, report or transaction to schedule this step using a process chain ?
Thanks
Christophe

The solution is to use a specific ABAP program using the method RESET_MODEL_RESULTS
You'll find BW-APD How to Reset a Data Mining Model" in the wiki
Christophe

Similar Messages

How to create info area for process chains?

How to create info area for process chains?

Hi,
If I understand your question clearly, you want to assign display componets.
once you have created you chain you assign the components by clicking CTRL+F11
Regards,
Namrata

How to Convert INFOPACKAGE GROUPS to PROCESS CHAINS???

Hi BW Gurus,
How to Convert INFOPACKAGE GROUPS to PROCESS CHAINS???and what are the
measures i need to take care before doing the same .
prasad

Hi,
Simply add these infopackage groups into process chain and schedule the process chain.
Hope this helps.
Assign points if useful.
Regards,
Venkat

How to identify whether that particular process chain is active or not?

Hi all,
How to identify whether that particular process chain is active or not ?
thanxs
haritha

If see the process chain then it will not be greyed out if it is active. Or else check in these tables.
RSPCCHAIN
RSPCCHAINT
to check the logs
RSPCLOGS
RSPCPROCESSLOG
Give the chain name and check in the object version.
Khaja

How to find out which meta process chain a local chain is involved in

Hello experts,
i would like to know how a local process chain is started by a meta chain. I have a local chain running every day. It is involved in a meta chain. How can i find out the meta chain in which the local chain is involved.
I appreciate every help i get.
Thank you all in advance.
Kind regards,
Ali

Hi everybody,
the question is answered.
Thanks and regards.
Ali

How to add Abap program in process chain to check the another process chain status and do accordingly

Hello All,
My requirement is related to process chain 1 . I have one process chain in that i have to put abap program in middle of that chain . This program will check the status of one another process chain 2 status ( whether it was completed or it is going on). if it is completed then continue the Process chain 1 to further activity else wait to complete Process chain 2 .
Here i have to use process chain 2 as a parameter.
Kindly provide me solution .
Regards
Saurabh

Hi Saurabh,
You need to create ABAP program in which you check the values for required process chain in
following table
RSPCLOGCHAIN (Feilds ANALYZED_STATUS LOG_ID)
Depend on values you can trigger next process or you can wait unitil this process chain complete.you need to check status from table using ABAP.
If you are not aware about how to write ABAP program then take help from ABAP Team.
Regards,
Ganesh Bothe

How to set delta load in process chain

hello all
The scenario is i created process chain for transaction data and all the infopackages are in initial load now I want to do delta,how to set delta in process chain,bcoz only the new records should load?
kindly anyone let me know.
regards
balaji

Hi Balaji,
In most projects thay dont have the "init" infopackages in the process chain. They have 2 infopackages for a load - One for init and the other for delta. Now that you have created the process chain what I suggested was to change the infopackage settings to delta load and change the description of the infopackage.
The other option would be to create a new infopackages for delta and include them in the chain replacing the "init" ones.
Bye
Dinesh

How to rollup cube data in Process chain?

I have loaded data into a cube, and when i look at the technical status it is green but there is nothing in the "request for reporting available"box. I went to Rollup Tab and manually execute the request. It worked.
But is there anyway that I can add a process type to do it automatically? I saw there are three process type related to Rollup, which one should I use?
1. Initial Fill of New Aggregates
2. Roll Up of Filled Aggregates/BIA Indexes (what does this mean?)
3. initial Activation and Filling of BIA Indexes
Thank you!

Dear experts,
We have the following problem: for certain figures we have on ODS level 2 daily updates: one at night and one at noon. During the week at noon we only load from the ODS into the Cubes only that data that is entered the same day (selection on CPU-Date in the InfoPackage). After loading the data at night we load the data which was entered the previous day (selection on CPU-Date in the InfoPackage). In this process we have a step in the process chain that deletes in the overlapping request of the previous day (which was loaded at noon).
Our Process Chain for loading the data from ODS into the cubes looks as following:
Delete Indexes of the cubes -> Load data from ODS into Cubes -> Generate Cube Indexes -> Delete overlapping Requests in InfoCubes
After filling the BIA-Indexes on this cube, the process chain gives an error message at the stage where the overlapping request should be deleted. The error messages says:
Uncaught Exception: Keyfigure is of type Float.
To solve this error at the moment we manually delete the BIA Indexes, delete the overlapping requests and fill the BIA Indexes again. Since the functions 'delete BIA Index' and 'Fill BIA Index' is not available in RSPC we can not do this automatically in the process chain.
I also tried like above mentioned taking the step 'roll up filled BIA Indexes' into the process chain, but the check of the process chain creates a message stating that i cannot do this when in the same process chain the steps 'Delete Indexes of the Cube' and 'generate Indexes of the cubes' are included.
Does anybody know a solution how i can delete overlapping requests in a process chain with filled BIA Indexes.
Many thanks in advance for your kind reply.
Best regards,
Ilja Dekeyser

How to redirect loading flow in process chain based on logic?

Hi Experts,
I have a scenario where I want to keep data for last 3 years in 3 different cubes.e.g. lets say cube 1 holds data for current 2006 year, cube 2 holds 2005 and cube 3 holds 2004.Now in next year 2007, I want to keep data for 2007, 2006 and 2005.I want to keep data for 2007 in cube 3 which was holding 2004 data.(delete 2004 data and load 2007).
This process should be automated i.e. which cube to load should be done auto.In short, <b>data for new cube should go into cube which is holding data for oldest year.</b>
I want to know :
1) Which Options I can use to do this?
2) What about ABAP program : custom table can be maintained to know which cube is holding what data, but how to redirect loading flow?
3) What about "Decision process type" in process chain?
4) Also would custom process type solve this functionality?
Any ideas would be highly appreciated.
Thanks in advance,
Sorabh

Hi Sorabh,
Its just an Idea, Im assuming that this would work. This should also work for INIT DELTA I guess. But would need proper testing.
Have a Custom Table ZCUBEYEAR and maintain the CUBE as the Key.
ZCUBEYEAR
CUBE     YEAR
Y1       2004
Y2       2005
Y3       2006.
In the update rule->Start Routine for Cube Y1, extract this entry from the table ZCUBEYEAR for Y1, which in this case would be 2004.
DELETE DATA_PACKAGE WHERE YEAR NE YEARFORCUBEY1.
in our case YEARFORCUBEY1 = 2004.
For cube Y2 and Y3 the Delete statement would be as follows in their Start Routines.
DELETE DATA_PACKAGE WHERE YEAR NE YEARFORCUBEY2.
DELETE DATA_PACKAGE WHERE YEAR NE YEARFORCUBEY3.
This would ensure that only 2004 data would flow to Y1, 2005 for Y2 and 2006 for Y3.
Once we come to a NEW YEAR, We need to run a program or Manually change the CUSTOM TABLE "ZCUBEYEAR" updating the cube Y1 with 2007, the Deltas would flow correctly.
Please email me at [email protected], we could have a detailed discussion.
Hope the above helps your cause.
Regards,
Praveen.

HOW TO STOP A SCHEDULING A PROCESS CHAIN IN SAP BPC

Hello Expert,
Can Any 1 tell me how to stop a scheduling of a process chain in sap bpc ??

Hi Neeraj,
Your question is not clear.
If you are asking - How to delete a schedule which runs a process chain/DM package from EPM, then please go to Data Manager -> click View Status arrow -> select view schedule status -> select the schedule, right click and choose delete schedule.
Neha

How to incorporate time in the Process chain messages

Hi,
I would like to create message after success or failure of process chain with time of failure in the subject.How to put system variable in the message subject?.
Your help will be appreciated.
thanks in advance.
Manju

raj,
thanks for your quick reply but i would like to incorporate the finish time in the mail subject.
for example mail subject is as follows.
"GL process chain has been finished at xx:xx time".
Here time (xx:xx) should automatically read and fill as per the finish time of the process chain.
hope now it is clear.any idea how it can be done?.
manju

How to change source system in process chain

Hi All,
In our Bw quality system, process chain is pointing to the Production system ,
I want to keep the process variant name same as production but want to change the source system
For example.. let us say currently process chains(All DTPs) are pointing to the Production server (PD001) ip (200.12.23.22)
now i want to keep the Server name same but want to point it to the quality system..so after change Process chain will show same name but it should point out to quality system.(let us say bwq 200.12.23.232).
Can you please help me on this

Hi Pravender,
In our BW system , we have same process chain in all the system (Dev, Quality and production)..and naming convention is also same ..(all process chain points to Production system only)..
So when I want to load data from development i just need to change the source system ip adrress to the Development Server (So names in process chain will indicate production but it will point to the development server), .
I want to know how can i change the source system mapping.
Hope i am clear now

How to put RSCRM_BAPI in the process chain?

Dear All,
I've used RSCRM_BAPI to extract data from a query into a table. These data in the table needs to be extracted again to the PSA in BW.
So I need to run the job defined in RSCRM_BAPI after the data has been loaded into cube and run the infopackage to extract data after the job has finished.
How can I put these actions in the process chain?
Thanks.
Bolun

Hello Bolun,
Yes you can schedule RSCRM_BAPI extracts in a process chain. You have to write a program that will invoke the RSCRM APIs. You can use the program below.
data error(50) type c.
data g_t_return    LIKE bapiret2 OCCURS 0.
data: l_proc        like rscrmmon-processed.
data: gv_batchid    like RSCRMMON-BATCHID,
      gv_jobname    like TBTCO-JOBNAME,
      gv_actstatus like TBTCJOB-STATUS,
      gv_jobcount   like TBTCO-JOBCOUNT,
      gv_dbstatus   like TBTCJOB-STATUS,
      l_status      type RSCRMBSTAT,
      l_wait        type i.
parameter: repuid TYPE rscrmrepuid obligatory.
parameter: extrname TYPE tabname obligatory.
parameter: repvar like RSCRMSTATUS-REPVAR.
parameter: overwrt like RSCRMCONTXT-ACTIVE default 'X'.
start-of-Selection.
*Open Query
CALL FUNCTION 'RSCRMBW_REPORT'
    EXPORTING
      i_mode      = 'OPEN'
      i_reportuid = repuid
    TABLES
      e_t_return = g_t_return.
if g_t_return is not initial.
    read table g_t_return TRANSPORTING NO FIELDS with key type = 'E'.
    if sy-subrc <> 0.
      concatenate 'Errors occurred when opening query' error into error.
      message error type 'E'.
      exit.
    endif.
endif.
*use this if you want pass a variant.
*Get variable values
*     CALL FUNCTION 'RSCRMBW_VARIABLE_POPUP'
*         EXPORTING
*               i_reportuid = repuid
*               I_VARIANT = repvar
*         TABLES
*               e_t_return = g_t_return.
*Execute Report
CALL FUNCTION 'RSCRMBW_REPORT'
    EXPORTING
      i_mode         = 'START'
      i_reportuid    = repuid
      i_execmode     = 'TABLE'
      i_extract      = extrname
      I_CLEAREXTRACT = overwrt
    IMPORTING
      E_BATCHID      = gv_batchid
      E_JOBNAME      = gv_jobname
      E_JOBCOUNT     = gv_jobcount
    TABLES
      e_t_return     = g_t_return.
if g_t_return is not initial.
    read table g_t_return TRANSPORTING NO FIELDS with key type = 'E'.
    if sy-subrc <> 0.
      clear error.
      concatenate 'Errors occurred when executing query' error into error.
      message error type 'E'.
      exit.
    endif.
endif.
if sy-subrc = 0.
    WHILE ( L_STATUS NE 'SUCC' ).
      CALL FUNCTION 'RSCRMBW_BATCH_STATUS_CHECK'
        EXPORTING
          I_BATCHID       = gv_batchid
        IMPORTING
          E_STATUS        = l_status
        EXCEPTIONS
          NO_JOB_DATA     = 1
          INHERITED_ERROR = 2
          OTHERS          = 3.
      IF l_status = 'RUNN' or l_status = ' ' or l_status = 'SCHE'.
        WAIT UP TO 30 SECONDS.
      ENDIF.
    ENDWHILE.
    IF l_status = 'SUCC'.
      if sy-subrc = 0.
        write 'Query finished'.
      else.
        clear error.
        concatenate 'Job ' gv_batchid ' failed.' into error.
        Message error type 'E'.
      endif.
    ENDIF.
endif.
*Close report
CALL FUNCTION 'RSCRMBW_REPORT'
    EXPORTING
      i_mode      = 'CLOSE'
      i_reportuid = repuid.
end-of-selection.
Use the ABAP Program process to execute this program. Create the data load processes so that they execute on successful finish of this program.
I hope this makes things clear.
Regards
Arvind

How can I realize a period process chain start: the 1st to 3rd day every?

Hi,All
I want to set a start in a process chain as the following:
a period value: the 1st to 3rd day every month
How can I realize it ? Thanks.
Best Regards
Jiande Ding

Hi Jiande Ding
create original chain as a Meta chain.
create the another process chain with immedaite in start varaint..
and put abap step in that PC after process chain..
put this below code in that abap program.
DATA: wwdate like sy-datum,
          WWDAY(2).
wwday = sy-datum+6(2).
IF WWDAY EQ '01' OR WWDAY EQ '03'.
   CALL FUNCTION 'RSPC_API_CHAIN_START'
EXPORTING
    I_CHAIN             = <ur meta chain tech.name>
ENDIF.
Edited by: shanthi bhaskar on May 5, 2009 4:44 PM

How to include Events Concept in Process Chain and how to trigger it

Hi
I have Three Process Chains PC1, PC2, PC3 .
I want to trigger there Process Chains by Events .
I want PC1 needs to run every day 7:00am . And once it is finished then PC2 needs to start . And once PC2 is finsihed then PC3 needs to execute .
So please give me step by step process for Process for implementing the Events in this model and give me the steps to include Events concept in my model .
Thanks Alot
Regards
rizwan

Hi,
You can do this without event also. Schedule PC1 first and then have PC2 as local to your PC1 and PC3 as local to PC2. In that way, as soon as PC1 is done it will trigger PC2 and as soon as PC2 is done it will in turn trigger PC3.
If you want it through an event, first create event in SM62 and you can trigger this event in SM64. Then for the PC1 in start variant, click on select conditions, select After Event and give this event name. Now again its better & efficient to trigger this even automatically, for which you need to write an ABAP program in SE38 and define a variant there. This ABAP program can be scheduled as a background job in SM36, click on Click on step on top. Give program name and variant.Click save.Next click on Start Condition and schedule is as your require it to run. So this background job starts automatically and runs your ABAP program which in turn will trigger your event and PC's will get triggered.
For both the approaches, you have to have PC2 & PC3 as Local chains as I explained above.
Guru

How to initialize Datamining Model in process chain ?

Similar Messages

Maybe you are looking for