Update Millions of records in a table

Hi ,
My DB is 10.2.0
This is a very generic question, hence I am not adding the Plan.
I have to update a column in a table with value of sum of another column from the same table.
The table has 19 million records.
UPDATE transactions a SET rest_amount = Nvl((SELECT Sum(trans_amount) FROM transactions b
WHERE b.trans_date=a.trans_date AND trans_ind < 2),trans_amount)
WHERE trans_grp < 6999 and trans_ind < 2;This query takes 10 hours to run. There are indexes on trans_date and trans_grp. There is no index on rest_amount column.
As per tom in
http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:6407993912330
He had suggested to disable all indexes and constraints and then enable them. Is it applicable only for INSERT or for all DML.
Only during insert the constraints will be validated and index will be refreshed with the new row. But in my case, I am updating a column which is not a part of any index. Should I go as per tom's suggestion ???

The table has 19 million records.How many rows (that's rows) are actually updated?

Similar Messages

How to Update millions or records in a table

I got a table which contains millions or records.
I want to update and commit every time for so many records ( say 10,000 records). I
dont want to do in one stroke as I may end up in Rollback segment issue(s). Any
suggestions please ! ! !
Thanks in Advance

Group your Updates.
1.) Look for a good group criteria in your table, a Index on it is recommend.
2.) Create an PL/SQL Cursor with the group criteria in the where clause.
cursor cur_updt (p_crit_id number) is
select * from large_table
where crit_id > p_crit_id;
3.) Now you can commit in a serial loop all your updates.

Best way to insert millions of records into the table

Hi,
Performance point of view, I am looking for the suggestion to choose best way to insert millions of records into the table.
Also guide me How to implement in easier way to make better performance.
Thanks,
Orahar.

Orahar wrote:
Its Distributed data. No. of clients and N no. of Transaction data fetching from the database based on the different conditions and insert into another transaction table which is like batch process.Sounds contradictory.
If the source data is already in the database, it is centralised.
In that case you ideally do not want the overhead of shipping that data to a client, the client processing it, and the client shipping the results back to the database to be stored (inserted).
It is must faster and more scalable for the client to instruct the database (via a stored proc or package) what to do, and that code (running on the database) to process the data.
For a stored proc, the same principle applies. It is faster for it to instruct the SQL engine what to do (via an INSERT..SELECT statement), then pulling the data from the SQL engine using a cursor fetch loop, and then pushing that data again to the SQL engine using an insert statement.
An INSERT..SELECT can also be done as a direct path insert. This introduces some limitations, but is faster than a normal insert.
If the data processing is too complex for an INSERT..SELECT, then pulling the data into PL/SQL, processing it there, and pushing it back into the database is the next best option. This should be done using bulk processing though in order to optimise the data transfer process between the PL/SQL and SQL engines.
Other performance considerations are the constraints on the insert table, the triggers, the indexes and so on. Make sure that data integrity is guaranteed (e.g. via PKs and FKs), and optimal (e.g. FKs should be indexes on the referenced table). Using triggers - well, that may not be the best approach (like for exampling using a trigger to assign a sequence value when it can be faster done in the insert SQL itself). Personally, I avoid using triggers - I rather have that code residing in a PL/SQL API for manipulating data in that table.
The type of table also plays a role. Make sure that the decision about the table structure, hashed, indexed, partitioned, etc, is the optimal one for the data structure that is to reside in that table.

How do I update all the records in a table from the contents of another table?

Ok some situation information, I have a pervasive database that runs our accounting software that I am pulling a product list from. I have that list stored in a table in SQL. From time to time we update the records in the pervasive database and
I want to be able to pull those changes into the SQL table. I don't want to drop the table and recreate it because if a product is no longer active in the pervasive database it will not be returned by the query (if I return all the products even the
inactive ones I will get thousands of records rather than just a few hundred) and I do not want to loose the products from the table that are now inactive.
So what I want to be able to do is pull the list from pervasive, compare it to the list that exists in SQL and update any changed records, add any new records, and leave any now missing records alone(missing from the pervasive list but present in the SQL
list). I have no trouble pulling the records from pervasive (now) but I am a little stumped on how to do the rest. I am not sure if this is a situation to use MERGE or not. I also do not really need this to be done on a regular basis as the
changes do not happen often enough for that, the ability to manually trigger it would be enough.
Any help would be appreciated.
David

Hi David,
lets say you want to go with the lookup transformation.
lets say u want to move the data from server A table A1 to Server B table B1
What you need to do is the following:
Cofigure the Lookup options as follows:
- In general -> Specify how to handle rows with no macthing entries -> "Redirect Rows to no Match Output"
- In Connection -> Set the ole db connection to the Server B and select the table B you want to lookup the values in your case the table where you want to input the changes
- In columns -> link the product column from table A to product column in table B. And do not select any rows to output.
- now your component is ready next you need to input. so when u connect your lookup to the destination component You will get an option to select which output you want to use - use "Lookup No Match Output".
this will actually just allow you to add only new items only to the table B.
Teddy Bejjani - BI Specialist @ Netways

Update Multiple DB records in one table with one form

I am developing an onine proofing system that will display a
number of images, and then allow the customer to approve each
image, as well as provide comments.
I currently have a table setup with the various elements (an
image that dynamically gets its name, details about the image, and
the neccessary form fields).
See This
Image for the layout
I would then apply a repeat region (the number of signs would
vary), and like to update all the records with the customer's
comments and approval with one form. How is this done? I have
downloaded a trial of the developers toolkit, but the documentation
is horrible. I would be willing to buy an extension that will allow
this, if one exists.

I found an extension that did this in about two minutes:
http://www.webassist.com/professional/products/productdetails.asp?PID=117
Excellent software, highly recommended. Easy to figure out,
and got this "problem" resolved fast.

Update the change record in z table

Hi,
I am doing a module pool programming.If i change the particular record only changed record shd be updated in the z table.
How can i update only changed record in the database table in module pool.
Regards

Hi Rasheed,
Use this code to update ztable from the module pool table control, its working.
At screen flow logic
PROCESS BEFORE OUTPUT.
MODULE STATUS_8001.
LOOP WITH CONTROL TAB.
    MODULE PASS_DATA. "module to pass data into table control
ENDLOOP.
PROCESS AFTER INPUT.
MODULE USER_COMMAND_8001.
LOOP WITH CONTROL TAB. "tab is the table control on the screen
    MODULE MODIFY_DATA. "module to modify data in internal table
ENDLOOP.
MODULE SAVE_DATA. "module to save the modified data into database table
Design a PF-Status with a SAVE button with function code 'SAVE'.
In PBO
MODULE STATUS_8001 OUTPUT.
SET PF-STATUS 'Z_TABCT'. "pf-status with appropiate buttons
* SET TITLEBAR 'xxx'.
DATA : LINE_COUNT TYPE I.
DESCRIBE TABLE ITAB
LINES LINE_COUNT.
TAB-LINES = LINE_COUNT + 3. "increment the line of table control dynamically so that table control becomes scrollable
ENDMODULE.                 " STATUS_8001 OUTPUT
MODULE PASS_DATA OUTPUT.
SORT ITAB.
READ TABLE ITAB INDEX TAB-CURRENT_LINE.
ENDMODULE.                 " PASS_DATA OUTPUT
In PAI
MODULE USER_COMMAND_8001 INPUT.
OK_CODE = SY-UCOMM.
CASE OK_CODE.
    WHEN 'BACK'.
      LEAVE TO SCREEN 8000.
ENDCASE.
ENDMODULE.                 " USER_COMMAND_8001 INPUT
MODULE MODIFY_DATA INPUT.
MODIFY ITAB INDEX TAB-CURRENT_LINE.
ENDMODULE.                 " MODIFY_DATA INPUT
MODULE SAVE_DATA INPUT.
OK_CODE = SY-UCOMM.
CASE OK_CODE.
    WHEN 'SAVE'.
      MODIFY ITAB INDEX TAB-CURRENT_LINE. "update modifed records
      UPDATE ZTABLE FROM TABLE ITAB. "update database table from internal table
ENDCASE.
ENDMODULE.                 " SAVE_DATA INPUT
Hope this solves your problem.
Thanks & Regards
Tarun Gambhir

Importing and Updating Non-Duplicate Records from 2 Tables

I need some help with the code to import data from one table
into another if it is not a duplicate or if a record has changed.
I have 2 tables, Members and NetNews. I want to check NetNews
and import non-duplicate records from Members into NetNews and
update an email address in NetNews if it has changed in Members. I
figured it could be as simple as checking Members.MembersNumber and
Members.Email against the existance of NetNews.Email and
Members.MemberNumber and if a record in NetNews does not exist,
create it and if the email address in Members.email has changed,
update it in NetNews.Email.
Here is what I have from all of the suggestions received from
another category last year. It is not complete, but I am stuck on
the solution. Can someone please help me get this code working?
Thanks!
<cfquery datasource="#application.dsrepl#"
name="qryMember">
SELECT distinct Email,FirstName,LastName,MemberNumber
FROM members
WHERE memberstanding <= 2 AND email IS NOT NULL AND email
<> ' '
</cfquery>
<cfquery datasource="#application.ds#"
name="newsMember">
SELECT distinct MemberNumber
FROM NetNews
</cfquery>
<cfif
not(listfindnocase(valuelist(newsMember.MemberNumber),qryMember.MemberNumber)
AND isnumeric(qryMember.MemberNumber))>
insert into NetNews (Email_address, First_Name, Last_Name,
MemberNumber)
values ('#trim(qryMember.Email)#',
'#trim(qryMember.FirstName)#', '#trim(qryMember.LastName)#', '#
trim(qryMember.MemberNumber)#')-
</cfif>
</cfloop>
</cfquery>
------------------

Dan,
My DBA doesn't have the experience to help with a VIEW. Did I
mention that these are 2 separate databases on different servers?
This project is over a year old now and it really needs to get
finished so I thought the import would be the easiest way to go.
Thanks to your help, it is almost working.
I added some additional code to check for a changed email
address and update the NetNews database. It runs without error, but
I don't have a way to test it right now. Can you please look at the
code and see if it looks OK?
I am also still getting an error on line 10 after the routine
runs. The line that has this code: "and membernumber not in
(<cfqueryparam list="yes"
value="#valuelist(newsmember.membernumber)#
cfsqltype="cf_sql_integer">)" even with the cfif that Phil
suggested.
<cfquery datasource="#application.ds#"
name="newsMember">
SELECT distinct MemberNumber, Email_Address
FROM NetNewsTest
</cfquery>
<cfquery datasource="#application.dsrepl#"
name="qryMember">
SELECT distinct Email,FirstName,LastName,MemberNumber
FROM members
WHERE memberstanding <= 2 AND email IS NOT NULL AND email
<> ' '
and membernumber not in (<cfqueryparam list="yes"
value="#valuelist(newsmember.membernumber)#"
cfsqltype="cf_sql_integer">)
</cfquery>
<CFIF qryMember.recordcount NEQ 0>
<cfloop query ="qryMember">
<cfquery datasource="#application.ds#"
name="newsMember">
insert into NetNewsTest (Email_address, First_Name,
Last_Name, MemberNumber)
values ('#trim(qryMember.Email)#',
'#trim(qryMember.FirstName)#', '#trim(qryMember.LastName)#', '#
trim(qryMember.MemberNumber)#')
</cfquery>
</cfloop>
</cfif>
<cfquery datasource="#application.dsrepl#"
name="qryEmail">
SELECT distinct Email
FROM members
WHERE memberstanding <= 2 AND email IS NOT NULL AND email
<> ' '
and qryMember.email NEQ newsMember.email
</cfquery>
<CFIF qryEmail.recordcount NEQ 0>
<cfloop query ="qryEmail">
<cfquery datasource="#application.ds#"
name="newsMember">
update NetNewsTest (Email_address)
values ('#trim(qryMember.Email)#')
where email_address = #qryEmail.email#
</cfquery>
</cfloop>
</cfif>
Thank you again for the help.

Inserting millions of records into new table based on condition

Hi All,
We have a range partitioned table that contains 950,000,000 records (since from 2004) which is again list sub-partitioned on status. Possible values of stauts are 0,1,2,3 and 4.
The requirement is to get all the rows with status 1 and date less than 24-Aug 2011. (Oracle 11g R2).
I trying below code
CREATE TABLE RECONCILIATION_TAB PARALLEL 3 NOLOGGING
AS SELECT /*+ INDEX(CARDS_TAB STATUS_IDX) */ ID,STATUS,DATE_D
FROM CARDS_TAB
WHERE DATE_D < TO_DATE('24-AUG-2011','DD-MON-YYYY')
AND STATUS=1; CARDS_TAB has tow global indexes one on status and another on date_d.
Above query is running for last 28Hrs! Is this the right approach?
With Regards,
Farooq Abdulla

You said the table was range partitioned but you didn't say by what. I'm guessing the table is range partitioned by DATE_D. Is that a valid assumption?
You said that the table was subpartitioned by status. If the table is subpartitioned by status, what do you mean that the data is randomly distributed? Surely it's confined to particular subpartitions, right?
What is the query plan without the hint?
What is the query plan with the hint?
Why do you believe that adding the hint will be beneficial?
Justin

Updation of Excise Record thru Direct Table Entries in Table J_2IACCBAL

Hi,
While updating Excise Opening Figures in Table J_2IACCBAL, I am able to make the following entries only,
RG23ABED
RG23ASED
RG23AAED
RG23CBED
RG23CSED
RG23CAED
PLABED
PLASED
PLAAED
PLACESS
and There is no field to enter Secondary Edn. Cess
How will the system updates These missing enteries.
This is quite urgent as we are in the GOLIVE stage
Thanks and Regards,
Raghu

Dear Raghu,
refer the note 1045888 for introducing SECESS in the table
J_2iaccbal.
For ECS do the following:
Here are the STEP BY STEP instructions...
for table J_2iaccbal
Go to se11
Put the table in database table
Change
select the field REGISTER/J_2IREGELE double click on J_2IREGELE.
You will then be able to see the Domain J_2IREGDOM
Double click on this J_2IREGDOM
You will then be able to see a tab called "Value range"
Here there is a column called "Fixed values", here you enter the following
"RG23AECS
RG23CECS
PLAECS "
Then you will be able to see the ECS elements als in the F4 function of
register field in J_2iaccbal table.
You need abaper help to made above change.
Regards,
Bhadresh

How to update the millions of records in oracle database?

How to update the millions of records in oracle database?
table have contraints & index.how to do this mass update.normal update taking several hours.

LostWorld wrote:
How to update the millions of records in oracle database?
table have contraints & index.how to do this mass update.normal update taking several hours.Please, refer to Tom Kyte's answer on your question
[How to Update millions or records in a table|http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:6407993912330]
Kamran Agayev A. (10g OCP)
http://kamranagayev.wordpress.com
[Step by Step install Oracle on Linux and Automate the installation using Shell Script |http://kamranagayev.wordpress.com/2009/05/01/step-by-step-installing-oracle-database-10g-release-2-on-linux-centos-and-automate-the-installation-using-linux-shell-script/]

How to update multiple records in a table created in view (web dynpro)

Here is my coding......
*coding to get the district value
DATA lo_nd_district TYPE REF TO if_wd_context_node.
    DATA lo_el_district TYPE REF TO if_wd_context_element.
    DATA ls_district TYPE wd_this->element_district.
    DATA lv_district_txt LIKE ls_district-district_txt.
navigate from <CONTEXT> to <DISTRICT> via lead selection
    lo_nd_district = wd_context->get_child_node( name = wd_this->wdctx_district ).
get element via lead selection
    lo_el_district = lo_nd_district->get_element( ).
get single attribute
    lo_el_district->get_attribute(
      EXPORTING
        name = `DISTRICT_TXT`
      IMPORTING
        value = lv_district_txt ).
*coding to diplay records when clicking a button(Submit)
DATA lo_nd_table TYPE REF TO if_wd_context_node.
DATA lo_el_table TYPE REF TO if_wd_context_element.
DATA ls_table TYPE wd_this->element_table.
DATA lv_district LIKE ls_table-district.
navigate from <CONTEXT> to <TABLE> via lead selection
lo_nd_table = wd_context->get_child_node( name = wd_this->wdctx_table ).
get element via lead selection
lo_el_table = lo_nd_table->get_element( ).
get single attribute
lo_el_table->set_attribute(
    EXPORTING
      name = `DISTRICT`
   " IMPORTING
      value = lv_district_txt ).
The above coding updates only one record to that
table created in view.
If i enter 2nd district value means then the first record
in the table is overwritten.
So my need is the record should not be overwritten.
it(2nd record ) should be displayed after the 1st record.
Any one can help me and send the coding plz....

instead of using set attribute you should use bind table method to display/update the records in table view.
step1 ) collect all the data in a local table
step2 ) and the bind that lacal table with your node
search1 = wd_context->get_child_node( name = `TABLE1` ).
search1->bind_table( lt_detail)
here lt_detail is your local table and TABLE1 is node which is bound with table ui element.

Updating only perticular records

Dear Members,
I have a problem with one update statement , before discussing my problem please looking to the tables' data as shown below
Table "hari" data is below
contract tot
12 0
13 0
14 0
15 5
Table "hari_tot" data is below
contract amount
12 1
12 2
13 1
13 3
14 1
14 4
15 1
15 5
Now please look into my update query
update hari a set a.tot=(select sum(b.amount) from hari_tot b where a.contract=b.contract and a.contract in (12,13))
Here i mean to update only contract id 12,13 details, but its going to update all the records in "hari" table data. So 4 updates are happening over here . Finally data will looks in "hari"
table as below
contract tot
12 3
13 4
14
15
To avoid null update i have modified the the above query as below
update hari a set a.tot=nvl((select sum(b.amount) from hari_tot b where a.contract=b.contract and a.contract in (12,13)),a.tot)
After updating the table "hari" result looks as follows
contract tot
12 3
13 4
14 0
15 5
Since i need to update only 100 or 150 records in a table which is having millions of records, if i ran the statements like above, i'm going to achieve my task but unnecessarily i'm going to update all the records. Because of this reason only I'm trying with small tables' data like above
Could you please let me know the query to update only 12,13 records details, i meant to say only two updates should happen if i run the query and the data should look as shown below
contract tot
12 3
13 4
14 0
15 5
Many Thanks,
Hari.

Hi,
solution was just under your nose:
UPDATE hari a
   SET a.tot =
          (SELECT SUM (b.amount)
             FROM hari_tot b
            WHERE a.contract = b.contract)
WHERE a.contract IN (12, 13)I don't mean to be rude but you were very closed to the correct solution.
As a suggestion for next posts read SQL and PL/SQL FAQ.
Additionally when you put some code or output please enclose it between two lines starting with {noformat}{noformat}
i.e.:
{noformat}{noformat}
SELECT ...
{noformat}{noformat}
It's quite useful if you could post CREATE TABLE and INSERT statement of a sample of your data.
Regards.
Al
Edited by: Alberto Faenza on Jan 9, 2013 3:40 PM

Insert a record into a table through email in an Oracle APEX application

I developed an Oracle APEX application, there is a table called events. I can insert/update/delete a record in the table through browser. I am thinking in order to quickly do the data entry, user should be able to send an email, then the table should be inserted with values from email address, timestamp, subject and body. Anyd idea how to realize this functionality?
- Denis

Start by checking whether your mail server provides any API's for accessing emails , if it does you might be able to reduce a lot of work by using some kind of web service consumer from apex to your mail server. In any case your implementation is going to be dependent on your Mail Server configuration.
Your problem breaks down to reading/accessing mails from the mail server from PLSQL (apex is driven by PLSQL).
I found this other thread which could be of some use.
WAY TO ACCESS A MAIL SERVER FROM ORACLE APEX
<li>The following package might solve your problem directly(from carsten czarski of the German Apex community)
[url http://plsqlmailclient.sourceforge.net]http://plsqlmailclient.sourceforge.net
PS: POP3 support is still TBD.
<li>I also found this posting in the orafaq forums which lists a java method and PLSQL code bit for it for accessing emails via POP3
[url http://www.orafaq.com/forum/t/80928/2/]http://www.orafaq.com/forum/t/80928/2/
If these do not work for you, find some java library to read mail from your server, write a PLSQL wrapper for it and use it in a scheduled job(DBMS_JOB)/a PLSQL block triggered from Apex.
If you get stuck there, find some utility that can read mails, invoke them from your DB using java,shell scrpt,dbms_scheduler etc and use the utility's function for the rest.
NOTE: I haven't tried any of these utilities and you must validate any java code before running them on your environment.
Since aren't really much restrictions(other than spam checks) in sending a mail to your mail account, you might want to consider filtering out the mails from which you create records.

Updating an existing data in the table...urgent

Hi All,
I am updating the existing records in the table.My CO piece of code is as below:
if(!vo.isExecuted() && !vo.isPreparedForExecution())
vo.executeQuery();
Row requestRow = vo.getCurrentRow();
requestRow.setAttribute("CustomerName", custName);
requestRow.setAttribute("CustomerNumber", custNumber);
but I am getting Null pointer exception at line:
vo.getCurrentRow();
Its very urgent.
Please help me on this.
Akshata
Edited by: Akshata on Apr 25, 2012 3:29 AM

Try below
if(!vo.isExecuted() && !vo.isPreparedForExecution())
vo.executeQuery();
Row requestRow = vo.first(); //If it also returns null, that means the VO doen't have any rows.
requestRow.setAttribute("CustomerName", custName);
requestRow.setAttribute("CustomerNumber", custNumber);
Regards,
Peddi.

Trigger to keep a single record form a table to a certain status

Hi all.
I have a table which has a "Status" field. I'm trying to accomplish, with a trigger, that the table can only have one record with an "Active" status, so whenever and Insert or Update occurs, if the new value for the "Status" field is "Active", I should update all other records in the table to "Inactive". I've tried several approaches, and managed to make it work on Inserts, but any Updates wll give an error that says that the table is mutating. I've tried to do it on before and after trigger types, without success.
Is there a way to do what I want with a trigger, or do necessarily I have to create an SP that needs to be called before the Insert or Update operation?

Alternatively you can try with as follows:
SQL> create table t as select rownum id, 'Inactive' status from dual connect by level <= 4
2 union all select 5, 'Active' from dual
3 /
Table created.
SQL>
SQL> select * from t
2 /
        ID STATUS
         1 Inactive
         2 Inactive
         3 Inactive
         4 Inactive
         5 Active
SQL> create or replace trigger t_trg
2     before insert or update
3     on t
4     for each row
5 begin
6     if (sys_context (
7                      'userenv',
8                      'client_info'
9              ) != 'start transaction'
10      or sys_context (
11                      'userenv',
12                      'client_info'
13              ) is null)
14     then
15             dbms_application_info.set_client_info ('start transaction');
16
17             execute immediate 'update t@loopback set status = ''Inactive'' where status = ''Act
ive'' and id != :id'
18                     using :new.id;
19
20             dbms_application_info.set_client_info ('end transaction');
21     end if;
22 end t_trg;
23 /
Trigger created.
SQL>
SQL> update t
2     set status = 'Active'
3   where id = 2
4 /
1 row updated.
SQL>
SQL> select * from t
2 /
        ID STATUS
         1 Inactive
         2 Active
         3 Inactive
         4 Inactive
         5 Inactive
SQL>
SQL> update t
2     set status = 'Active'
3   where id = 4
4 /
1 row updated.
SQL>
SQL> select * from t
2 /
        ID STATUS
         1 Inactive
         2 Inactive
         3 Inactive
         4 Active
         5 Inactive
SQL>
SQL> insert into t values (6, 'Active')
2 /
1 row created.
SQL>
SQL> select * from t
2 /
        ID STATUS
         1 Inactive
         2 Inactive
         3 Inactive
         4 Inactive
         5 Inactive
         6 Active
6 rows selected.
SQL>All you need is the loopback trigger wich I defined as
CREATE DATABASE LINK LOOPBACK
USING '(description=(address=(protocol=beq)(program=/path_to_your_oracle_binary/oracle)))';

Update Millions of records in a table

Similar Messages

Maybe you are looking for