How to Improve Delete Process

How to Improve Delete Process for File Reloads of Same Month of Data on Cubes.We currently use a SAP generated program for selective deletion to do this. It would be more efficient to delete by InfoPackage request (or packet) and use delivered processes in process chains to do this.How can we use the more efficient delete processes in the process chains to perform the task.Could u explain me with steps as it would be more clear

Similar Messages

How to find deleted process chain and query

Hi All
Some body has deleted one process chain and one query from development system
how can i know who has deleted these?
is there any table where deleted process chain and queries will maintain????
Thanks & Regards,
Krishna
Edited by: krishnamurthy g on Sep 1, 2008 5:33 PM

Hi,
Usually its totally in tractable to find a deleted chain :
But you can try out with the following options:
1> T-code SLG2 .
2> T-code SLG1 - find out who was logged on to the system on that particular client at that time.
3> Try out the following tables in se16 :
RSPCLOGS --- Application Logs for the Process Chains
RSPCLOGCHAIN --- Cross-Table Log ID / Chain ID
RSPCCOMMANDLOG --- System Command Execution Logs (Process Chains) .
Hope it will be expedite.
Thax & Regards.
Vaibhave Sharma

How can i delete processing podcasts?

I've been waiting two weeks for my podcasts to finish processing.How can i delete my podcasts after they start processing?
I really need help! Thx:)

If you have another download that did not complete - Let it finish or try to delete it as well. If its a song, finish the download or the the swipe to delete feature, it its an app try tapping and holding down on it to delete it or tap on the icon to finish the download.
Eiether complete those downloads or try to delete them and then see if you can use the swipe to delete or tap and hold down on the podcast to delete them.
You can also try Settings>General>Music>Podcasts and see if you can delete them in there. Tap the edit button and see if you get the red circle to the left of the podcast.

How to improve the process chain loading time

hi gurus,
one process chian having start - loading - roll up (3activities) but the load is data mart load which is going to update 11further targets. this take 1hrs. we want reduce this load. pl give ideas about reducing process chain time. (there are no indexing in the chain)

delete index and re-create after loading will surely improve loading performance but only it is only applicable if you do not compress the data , because if you compress the data will be available in E fact table.
Check the tranfer and update rules, if you have routines that read tables, check those statements and see if they can be still more optimized.
In the infopackage see ifyou have load PSA and then in to data target , if so check if there is any possibility of changing to load both in parallel.
hope it is clear
assign points if useful

How to speedup delete process.

Hi All,
In our production database(ORACLE 9.2.0.7), we are having a table of 45GB having 40crore records(this includes 60 stores items).
Now they are planning to delete the items which are of not useful.
So they are runniing query like DELETE FROM <TABLE_NAME> WHERE STROE=<STORE_ID>.
They are complaining like it is taking more than 3 hours per store.
My question, is there any way to speed up this process.
Thanks a lot.

user615922 wrote:
Hi All,
In our production database(ORACLE 9.2.0.7), we are having a table of 45GB having 40crore records(this includes 60 stores items).
Now they are planning to delete the items which are of not useful.
So they are runniing query like DELETE FROM <TABLE_NAME> WHERE STROE=<STORE_ID>.
They are complaining like it is taking more than 3 hours per store.
My question, is there any way to speed up this process.If you have triggers, may be enabled cascaded foreign keys in place and the operation needs to be online because the table is modified by other sessions during the delete, there is not much you can do apart from checking the execution plan and tracing the session where most of the work is done.
But if you can afford to do this as an "offline" operation, you might want to refer to Jonathan Lewis' "Big Update" post. The same basically applies to your case.
http://jonathanlewis.wordpress.com/2008/07/28/big-update/
Note that there one further option that is only mentioned in the comments of the note which is based on an exchange partition operation. You would create a partitioned table, e.g. using Create Table As Select ... NOLOGGING (this way you prevent the generation of UNDO and REDO, but be aware of the implications of NOLOGGING with regards to recovery and stand-by databases) and the query to create the table already filters out all the records that you want to have deleted.
Afterwards you can run a ALTER TABLE EXCHANGE PARTITION WITH TABLE <your_table> and the contents are swapped. If you're satisfied with the results, drop the partitioned table, otherwise, undo the operation by repeating the EXCHANGE operation.
Note that this approach does have numerous restrictions (your triggers won't fire, you need to be careful with any indexes, etc., see e.g. here: http://download.oracle.com/docs/cd/B28359_01/server.111/b28286/statements_3001.htm#BABBBDDJ) and is only a sensible strategy if you don't have concurrent DML activity.
The nice thing about this approach is that from a reader's perspective the operation is "seamless" which means you can have on-going read-only access on the object, because queries that started before the exchange is performed will continue to read from the old segment whereas queries that start after the exchange will see the new data.
Regards,
Randolf
Oracle related stuff blog:
http://oracle-randolf.blogspot.com/
SQLTools++ for Oracle (Open source Oracle GUI for Windows):
http://www.sqltools-plusplus.org:7676/
http://sourceforge.net/projects/sqlt-pp/
Edited by: Randolf Geist on Oct 28, 2008 10:02 AM
added the caveat regarding NOLOGGING

How to resume deleted process chain

HI BW gurus,
I have 1 process chain which loads data to ZFIGL_01 infocube.But by mistake the process chain got deleted from the meta chain as well as main chain.
Is there any step to recreate the process chain.And will it affect my main process chain which will run tomorrow at 3.30
Please guide.
Regards,
Shailja kaul.

You can go ahead and create the process chain and add it to your main process chain also. After you have done that. Make sure that you activate and schedule it.
After scheduling the Process chain you can go ahead and check for the scheduled job by right click on the start process.
Let me know if you face any problems
Regards,
Abhishek

How to improve DELETE statement that remove millions of rows?

The following query take lot of time when exectued, even after I drop the indexes, is there a better way to write the following query?
DELETE from pwr_part
where ft_src_ref_id in (select ft_src_ref_id
from pwr_purge_ft);
--Table:pwr_part
--UIP10371 foreign key (FT_SRC_REF_ID, FT_DTL_SEQ)
--Table: pwr_purge_ft
--PWR_PURGE_FT_PK Primary key (FT_SRC_REF_ID, FT_DTL_SEQ)
select count(*) from pwr_part;
--27,248,294
select count(*) from pwr_purge_ft;
--23,803,770
Explain Plan:
Description Object owner Object name Cost Cardinality Bytes
SELECT STATEMENT, GOAL = ALL_ROWS 224993 5492829 395483688
HASH JOIN RIGHT SEMI 224993 5492829 395483688
INDEX FAST FULL SCAN PWR_OWNER PWR_PURGE_FT_PK 43102 23803770 142822620
PARTITION HASH ALL 60942 27156200 1792309200
TABLE ACCESS FULL PWR_OWNER PWR_PART 60942 27156200 1792309200

Helio Dias wrote:
Have you ever thought about bulk collection?
http://heliodias.wordpress.com/2010/01/07/best-way-to-delete-millions-rows-from-hundred-millions-table/
One reason for which I would hate your suggestion.
Regular Delete vs Bulk Delete

Improve speed for deleting process

Hi all,
I am writing a housekeeping program to check on the table ZTP_ADA for the invalid date format.
If invalid date format record is found, then delete the record (delete those records which are erroneous). The table is having a huge number of records, might reach around millions of records.
To make it short, the program will does this: -
1) Delete records which are in invalid date format.
My question is, how to improve the speed for the current program, which might takes more than half an hour for deleting records.
Any suggestions or ideas are greatly appreaciated.
Million thanks.
DATA: lv_datum TYPE sy-datum,
      lv_commit TYPE c,
      lv_excep TYPE REF TO CX_ROOT.
DATA: ITAB_ADA TYPE SORTED TABLE OF ZTP_ADA WITH NON-UNIQUE KEY TABLE LINE,
      WA_ADA LIKE LINE OF ITAB_ADA.
SELECT * from ZTP_ADA
INTO TABLE ITAB_ADA.
LOOP AT ITAB_ADA INTO WA_ADA.
lv_datum = WA_ADA-DATUM.
CALL FUNCTION 'DATE_CHECK_PLAUSIBILITY'
EXPORTING
date = lv_datum
EXCEPTIONS
plausibility_check_failed = 1
OTHERS = 2.
IF sy-subrc <> 0.
DELETE FROM ztp_ada
    WHERE DATUM = WA_ADA-DATUM.
ENDIF.
ENDLOOP.

Did u try to remove the sorted table option from the internal table.Sorted table needs to be used only in case the records are sorted. And it adds to the processing if it is not sorted properly.
Make the internal table Standard table
Also I find that you are actually deleting the entries from the database table.
Delete the entries from the internal table which are having the wrong date format.
Then delete all the entries from the data base table and update the internal table into the database. This will be faster and you will hit the data base only thrice in the whole program.
This will fasten the process.
I would have used the following code below
DATA: lv_datum TYPE sy-datum,
      lv_commit TYPE c,
      lv_excep TYPE REF TO CX_ROOT.
DATA: ITAB_ADA TYPE STANDARD TABLE OF ZTP_ADA,
      WA_ADA LIKE LINE OF ITAB_ADA.
SELECT * from ZTP_ADA
INTO TABLE ITAB_ADA.
DELETE * FROM ZTP_ADA
INTO TABLE ITAB_ADA.
LOOP AT ITAB_ADA INTO WA_ADA.
lv_datum = WA_ADA-DATUM.
CALL FUNCTION 'DATE_CHECK_PLAUSIBILITY'
EXPORTING
date = lv_datum
EXCEPTIONS
plausibility_check_failed = 1
OTHERS = 2.
IF sy-subrc 0.
DELETE ITAB_ADA WHERE DATUM = LV_DATUM.
ENDIF.
ENDLOOP.
INSERT ZTP_ADA FROM TABLE ITAB_ADA.
Hope this helps.
Regards,
Hari

H how to find if the psa table is included in psa deletion process chain

Hi all
can anybody tell me .
I have a psa table it's technical name is /BIC/B*
h how to find the above psa table is included in psa deletion process chain or not
please help me

Hi
Ope the PC in RSPC1 -->go to planning view of process chain
in left side you will find different types of process types.
under other BW process types folder -->you will find process type "Delete request from PSA"
drag this into your process chain planning view and customize it based on your requirements.
check the below article for step by step procedure
http://www.sdn.sap.com/irj/scn/go/portal/prtroot/docs/library/uuid/a02ba9e7-bb6f-2c10-09b4-e86b9fcbad41?QuickLink=index&overridelayout=true
Regards,
Venkatesh

I inadvertently created a very large file. Not needing it I sent it to the Trash Can. However, the deletion process never completes. Any suggestions as to how to delete it ?

I inadvertently created a very large file on my hard drive. Not needing it I sent it to the Trash Can. However, the deletion process never completes. Any suggestions as to how to delete it ? I re-installed the OS but the file was still there taking up needed space.

Command-click all of the files you want to delete and don't actually move them to the Trash until you've gone through the entire folder.
(63341)

How to delete Process Variants in Process chain

Hello,
I created few test Process Variants I would like to delete how can we delete those?
Thanks,

I created few process variants but i did not saved those in process chain so i can not edit those in process chain
when I am trying to create new process chain in the popup window still showing old variants how to delete those
Thanks,

How to delete process forcedly

There are some problems in BPM Standalone 10g. The workspace shows a process is not available. I try to undeploy it from the BPM standalone. However, it cannot be undeployed.
I just changed the IP address one hour ago.
Error message:
The Process '/Process#Default-1.0' is not available. Caused by: Process '/Process#Default-1.0' not available. Caused by: Engine 'bpmengine' cannot process requests.
The bpmengine cannot be re started
Does anyone know how to delete process project directly?

I change the IP address to the original one. It works.

How to reuse the process chain which has been deleted?

hi
guys
how to reuse the process chain which has been deleted?
your regards
suresh

if you want to "recover" the deleted process chain, try to check DEV / QAS / PRD. Maybe there is a copy still available
Or check the following tables in SE11: RSPCCHAIN / RSPCCHAINATTR
Good luck !

How to Delete Process Chain?

How to Delete Process Chain?

When in the change mode of a process chain, choose the menu entry 'Process Chain --> Delete' (Only available on change mode).
Cheers
SAP NetWeaver 2004s Ramp-Up BI Back Office Team

How to delete process flow jobs

hi all,
need a piece of advice here. any idea how i can delete a process flow job scheduled using OEM without?? is there another way to do it without having Oracle Workflow Monitor? plz advice..
regards,
ykl

Hi,
If you meant to delete all the Process Flows deployed into Work Flow schema without using OEM, then take a look at the scripts available in WF Schema under the directory - <your oracle home>/wf/admin/sql.
The scripts you will be interested in are wfrmitt.sql which prompts you for the Process Flow Package Type and removes all the relevant information from the Work Flow schema. There are other scripts as well. Please look into those.
Thanks
Mahesh

How to Improve Delete Process

Similar Messages

Maybe you are looking for