Procedure of ODI with slow performance (SOURCE and TARGET are different Oracle databases)

Similar Questions

• Source and target tabs in the procedure

  I'm trying to create a procedure to insert data into a table.
  
  My order on the tab target:
  
  INSERT INTO insert_table
  
  My order on the Source tab:
  
  SELECT BATCHNUMBER,' it, COUNT (*) IN table_source WITH BATCHNUMBER, 'C '.
  
  I selected the logic diagrams correctly in the source and target tabs. I get the following error...
  
  
  java.sql.BatchUpdateException: ORA-00926: lack of keyword VALUES
  
  What would my orders in the target and Source tabs?
  
  -app

  For the string type data please include in single quotes, for example, if C is varchar2

  INSERT INTO TRGT_TABLE VALUES (#BATCHNUMBER,'#C',#COUNT) 
  

  Please try this

• AD, used as a trusted source and target system

  Hello

  I have a requirement to use AD as a trusted source and the target system the zero days and 1 for a transitional period.  We have 18 applications to integrate with OAM for SSO.  Currently, the authentication and authorization of these applications are made via AD.  However, the client wants to move to the use of LDAP in goal SSO.  The first phase includes 3 apps on the 18 apps.

  Day zero, I use AD as a source of confidence to push users to IOM.  Then, I run AD as a target system to link the user to their existing AD accounts.  Their ad groups will also be reconciled by IOM.

  Because there will be a transition period and the customer would not have to change the process of assistance to creating accounts AD (for internal and external users), they asked that we continue to allow accounts AD to reconcile with IOM in the trusted source and target system.

  I have not used AD connector as a reliable source.  I intend to source AD done trust reconciliation to run first, then the task of reconciliation system AD target to run then.  It's the same AD connector.

  This can work as long as the customer wishes until they want to spend to IOM for the creation of the user and accounts AD commissioning through the access policy.

  Is this sound ok you?  There is a "witch hunt" I didn't think?

  Thank you

  Khanh

  Yes, you can run the recon trust first and get all the identity created by IOM.

  Later, you can run the Scheduler for recon target for linking AD accounts with user profile of IOM.

  Note: There is a field in AD IT resource in need of update if you want to switch between the target and recon trust.

  Research of configuration:

  This parameter contains the name of the lookup definition that stores configuration information used during the reconciliation and commissioning.

  If you have set your target system as a resource target, then enter Lookup.Configuration.ActiveDirectory.

  If you have set your target system as a reliable source, then enter Lookup.Configuration.ActiveDirectory.Trusted.

  Default value: Lookup.Configuration.ActiveDirectory

  ~ J

• Database of waiting on a Source and target different endian formats

  Hello
  
  I have one familiar with the case where the source and target servers are different endian formats. I want to implement a 11g Data Guard in this environment. How can I do to implement the problem of different endian formats.
  
  Thank you

  As indicated in the documentation you quote, Data Guard is not possible when the main and backup databases are on platforms with a different endian format. If the endian format is the same, heterogeneous platforms may, or may not, be supported. The final documentation seems to be MOS Note 413484.1 (for physical standby) and 1085687.1 (for logical standby).

• TABLE import in the source and target OLAP and OLTP Informatica records

  In the designer to map every time I import from my source OLTP and OLAP source table is displayed in its own folder named instance imported from. It is a problem that when I migrate my lower the workflow repository is always looking ofr these sources and targets. How to import or migrate there relative OLTP and OLAP source and target material in the table designer.
  
  Appreciate the help.

  Hi, before you migrate you can choose options as explained below

  1. create the same global name of connection for ODBC OLAP and OLTP and then import the tables to informatica

  2. other is, after you import the source tables with any ODBC connection name, you can change the name of the file as below
  a. check the table and in the Source Analyzer workspace to modify the table. In the table , tab click on rename
  (b) change the name of the data base by "OLAP" or 'OLTP' depending on your source.
  c. the source table is automatically moved to the folder OLAP and OLTP

  3. If the table is already existing in OLTP or OLAP, you can use the reuse or replacement of options when importing new mappings during the migration.

  Just to test this scenario once and applies for all tables in the source. I do the same during the migration

  Hope this helps

• Is it possible to have the source and target schema in the same instance of DB?

  Hi all
  
  I'm using Oracle 11 g 1 material.
  I spent another source than with OWB server locations.
  In the course of deploy I get VLD-3064 and I can't deploy mapping due to the many warnings "table or view does not exist.
  
  Is it possible to have the source and target schemas in the same case?
  How to do?
  
  Kind regards
  Martin

  Hi Martin!

  1. the target schema have select rights for source-tables/views.
  (Run as a user with dba rights: grant select on to ;).

  2 «.. . none generated code will use the link dataabase...'.
  This is only a warning and means there is no need to use a database link. If your mapping will be executed faster as using a database link.

  error of VLD 3064

  Greetings
  Guenther Herzog

• Fields "Synonym of entity source" and "Target entity synonym" for?

  Hello

  He is a newbie question.

  Is someone can you please tell me what that fields "Entity synonym Source" and "Synonym of target entity" in the dialog box properties of relationship are for? They are drop down boxes and look as if you should be able to select from the list of synonyms put in place for the entity in the properties of the entity dialog box.

  Thanks in advance.

  John

  Using SQL Developer Data Model Version 4.1.1.888 on Windows 7 Home premium 64-bit.

  Hi John,.

  If you right-click on an entity on the diagram object, then select Create Synonym in the drop down below, another entity object will be added to the diagram of the same entity.

  The fields "Synonym of entity Source" and 'Target entity synonym' in the relationship properties dialog box allow you to specify which of these multiple representations is connected to the relationship on the diagram.

  David

• FDM to connect to a relational source and target

  I'm new to this tool. I have a few basic questions. I based my research, but could not find a clear answer.

  Can I connect to a relational table as source and target for an application of FDM. Relational source isn't eBS or other applications, just a relational table. I would like to load the same data from a relational table to another table using FDM. I'm looking at the wrong tool? Should which adapter I use? ERPI source adapter is? Which adapter target?

  Can someone please you suggest? Thanks in advance.

  Hello

  If you're new to this tool, you should know that FDM won't 11.1.2.4.

  It is replaced by FDMEE which is already available in 11.1.2.3.

  In all cases, you can extract data from relational db by using a script of integration. FDM doesn't load data to a target database not Hyperion EPM but you he could get it works using the custom script.

  Concerning

• Schema (source and target)

  Hello
  
  I have a basic question. now I'm in the schema of the Designer tab, and I don't have an idea where I need to drag the source and target
  
  Thanks for the support
  
  Sam

  Hi Sam,

  I guess that you had created the following before you start your 'diagram' of Interface

  1. a project
  2 models (according to your requirement)

  Now, after you had provided information to the "definition" of the interface. In your diagram, you can goto "models" and then drag and drop the required for the models that you created in step 2 above. At the time where you drag and source and target, you can also auto card

  Hope this helps

  Sandeep Reddy, Enti
  HCC
  http://hyperionconsultancy.com/

• Query metadata ODI to find the source and the target table for Interface

  Hi Experts,
  
  Customer heading there source of EBS 11.5.10 in R12. They are BIApps ODI. 7952 version. Since then, all mappings there are customized they may not disturb on support of Oracle BIApps is concerned.
  Now, we need to know how many ODI mappings there source EBS is the 11.5.10 in R12 migration and we can only target mappings accordingly.
  
  So, please give me with entries below:
  
  (1) any request for metadata that will give me the information of the source tables table and target them in against an if-same interface that the source of the main interface is another interface.
  (2) what are the other stuffs that I need to look at from the point of view of the mapping changes when the source is upgrading.e.g. only change the source table is enough or I need to focus on other animals. I feel she's boiling down to create a separate source for R12 adapter.
  
  Kind regards
  Snehotosh

  First open the designer-> models and check which model has got the mapping of the eBS. Get this code with this TERRIBLE application

  SELECT count (distinct i_pop) of SNP_POP_COL
  where I_POP_COL in
  (
  Select i_pop_col in the SNP_POP_MAPPING where i_source_tab
  in
  (
  Select i_source_tab in the SNP_SOURCE_TAB where i_table in)
  Select i_table in the SNP_TABLE where I_MOD in
  (select i_mod in the SNP_MODEL where mod_name = "PUT_HERE_YOUR_NAME")
  )
  )
  );

  Will not consider some other influence (i.e. the procedure), it will be an indicator of average first level.

• Source and target data store mapping query

  I have to get the source and mapping target in ODI interface.

  Which table will I hit to get mapping information.

  E.g.

  Interface: INT_SAMPLE

  Data store: Source_DataStore with columns (cola, colb, teachers) Target_DataStore with columns (cola, colb, cold)

  Well mapping cover the QuickEdit tab and expand the field of mapping mapping is so

  Source_DataStore.Cola = Target_DataStore.Cola

  Source_DataStore.colB = Target_DataStore.colB

  
  

  
  

  Now, I want to get mapping information above as well as the name of the interface and the rest of the column that are not mapped using SQL (is it possible to trick ODI for mapping).

  Hi Prashant da Silva,

  Are you looking for an application to run on the repository?

  If so, it can help:

  select I.POP_NAME INTERFACE_NAME, ds.ds_name DATA_SET
        , s.lschema_name SOURCE_SCHEMA, NVL(S.TABLE_NAME, S.SRC_TAB_ALIAS) SOURCE_TABLE
        , mt.lschema_name TARGET_SCHEMA, I.TABLE_NAME TARGET_TABLE, c.col_name  TARGET_COLUMN, t.FULL_TEXT MAPPING_CRITERIA
    from SNP_POP i, SNP_DATA_SET ds, SNP_SOURCE_TAB s, SNP_TXT_HEADER t, SNP_POP_MAPPING m, SNP_POP_COL c, SNP_TABLE trg, snp_model mt
    where I.I_POP = DS.I_POP  (+)
      and DS.I_DATA_SET = S.I_DATA_SET (+)
      and T.I_TXT (+) = M.I_TXT_MAP
      and M.I_POP_COL (+) = C.I_POP_COL
      and M.I_DATA_SET = DS.I_DATA_SET (+)
      and C.I_POP (+) = I.I_POP
      and I.i_table = trg.i_table (+)
      and trg.i_mod = mt.i_mod (+);
  

  Just add a filter on UPPER (I.POP_NAME) = UPPER ('').

  Kind regards

  JeromeFr

• Partition member in source and target must be the same

  Hello gurus, I have a cube with partitions and each partition is to have about 10 dimensions.

  A dimension, I became a member and wrote a rule for this member in the partition of the source, I have to take the same Member in the score target to write the rule.

  No, you don't have to have the same set of members on both sides.

  If you have...

  A, B, C, D

  .. .in the source, and...

  A, B, C

  .. .in the target, you have two options.

  First option, live with the warnings of validation "incompatibility of cell count" and do nothing to do.  Then no data will appear in the target for had 'and had data' will be not only be pushed in any of the partition.

  Second option, card would be "one of the 'A', 'B' and 'C'."

• How can I change the connections of source and target OWB?

  Hello, I have hurt to change the connection to the source in OWB.
  
  My login name of source is identical to the front. That DB name, schema name is now moving to connect to the source. Here are the steps that I did.
  
  1. went to control Center Manager and cancel the mapping of source.
  2. is go explore connection and right click on the connection to the source, and then click Open Editor. change the username, ip address, password, name of the comic.
  3. goes to control Center Manager and redeploy the mapping. It does not work... It is said, table or view does not exist.
  
  Is there somewhere else I need to change... Somewhere, Miss me...
  
  Any help is appreciated...

  Hello

  Go to the module using the mentioned location.
  Change the module, set the location of metadata to the correct location
  Go to data locations, the correct location should be with selected locations. If so, remove it and place it again.

  Now, go to the configuration of the module and go to the Identification
  Check the value of the location and make sure that it uses the correct location

  I think it is sometimes useful to change the location to another location and back to the correct location.
  I hope this helps.

  Kind regards

  Emile

• MBRC and SYSTEM STATISTICS in Oracle database 10g.

  Hi all
  
  I am performing the upgrade of Oracle 8i Solaris database to Oracle 10 g HP - UX using the exp/IMP.
  
  But I have some doubts about the MBRC and system statistics.
  
  MBRC in Oracle 10 g is adjusted automatically if MBRC parameter is not defined, but I found the value 128 as shown below.

  SQL> select * from v$version;
  
  BANNER
  ----------------------------------------------------------------
  Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - 64bi
  PL/SQL Release 10.2.0.4.0 - Production
  CORE    10.2.0.4.0      Production
  TNS for HPUX: Version 10.2.0.4.0 - Production
  NLSRTL Version 10.2.0.4.0 - Production
  
  SQL> sho parameter multi
  
  NAME                                 TYPE                             VALUE
  ------------------------------------ -------------------------------- ------------------------------
  db_file_multiblock_read_count        integer                          128

  Also, I made a simple full table scan to test... but read scattered file db performs on 128 blocks. So I don't think that 128 and is automatic, I mean MBRC is not defined in consequence he always uses 128.
  
  The MBRC value affects the performance of the whole base?
  
  About SYSTEM STATISTICS, I found below result:

  SQL> select * from AUX_STATS$
  SNAME                          PNAME                               PVAL1 PVAL2
  ------------------------------ ------------------------------ ---------- --------------------------------------------------
  SYSSTATS_INFO                  STATUS                                    COMPLETED
  SYSSTATS_INFO                  DSTART                                    11-09-2009 04:59
  SYSSTATS_INFO                  DSTOP                                     11-09-2009 04:59
  SYSSTATS_INFO                  FLAGS                                   1
  SYSSTATS_MAIN                  CPUSPEEDNW                     128.239557
  SYSSTATS_MAIN                  IOSEEKTIM                              10
  SYSSTATS_MAIN                  IOTFRSPEED                           4096
  SYSSTATS_MAIN                  SREADTIM
  SYSSTATS_MAIN                  MREADTIM
  SYSSTATS_MAIN                  CPUSPEED
  SYSSTATS_MAIN                  MBRC
  SYSSTATS_MAIN                  MAXTHR
  SYSSTATS_MAIN                  SLAVETHR

  Now if it is better NOWORKLOAD or workload, and this server is still in the process of building... so how can I accumulate the workload statistics, as it is not possible to make a high load on this server? Is he really need to collect system statistics, what will happen with stats NOWORLOAD?
  
  I saw no single database where system stats are gathered in our organization with more than 2000 databases.
  
  -Yasser

  Perhaps what this article written by Tom Kite help:

  http://asktom.Oracle.com/pls/asktom/f?p=100:11:0:P11_QUESTION_ID:499197100346264909

• How to set up a query - 'A-lines' and 'E' are different.

  Hi all
  
  I use 11 g R2 and I'm trying to resolve a query. On the execution below plan, my E-lines and my A are different. As you can see, there is little the same line 22 to 43, but starting from line 21 to 1 E-lines are false. For example, on the content (RowSource) 12, E-lines is (1 * 179) and lines A is 8631.
  
  All statistics have been generated and are up to date. Do you know what could be the problem?
  

  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  | Id  | Operation                                     | Name                      | Starts | E-Rows | A-Rows |   A-Time   | Buffers | Reads  | Writes |  OMem |  1Mem |  O/1/M   | Max-Tmp |
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
  |   0 | SELECT STATEMENT                              |                           |      1 |        |     50 |00:05:55.08 |     203K|  69975 |   1094 |       |       |          |         |
  |   1 |  SORT ORDER BY                                |                           |      1 |    179 |     50 |00:05:55.08 |     203K|  69975 |   1094 |  8307K|  1133K|          |         |
  |   2 |   HASH UNIQUE                                 |                           |      1 |    179 |   8631 |00:05:54.56 |     203K|  69317 |    180 |  4474K|  1861K|          |   16384 |
  |   3 |    NESTED LOOPS                               |                           |      1 |        |   8631 |00:05:54.04 |     203K|  69137 |      0 |       |       |          |         |
  |   4 |     NESTED LOOPS                              |                           |      1 |    179 |   8631 |00:05:23.24 |     194K|  63768 |      0 |       |       |          |         |
  |   5 |      NESTED LOOPS                             |                           |      1 |    179 |   8631 |00:05:13.51 |     177K|  61396 |      0 |       |       |          |         |
  |   6 |       NESTED LOOPS                            |                           |      1 |    179 |   8631 |00:04:03.74 |     151K|  52928 |      0 |       |       |          |         |
  |   7 |        NESTED LOOPS                           |                           |      1 |    179 |   8631 |00:03:32.15 |     125K|  46425 |      0 |       |       |          |         |
  |*  8 |         HASH JOIN                             |                           |      1 |    179 |   8631 |00:02:59.42 |   99470 |  40095 |      0 |  4379K|  1869K|     1/0/0|         |
  |*  9 |          HASH JOIN                            |                           |      1 |    179 |   8631 |00:02:59.15 |   98919 |  40067 |      0 |   909K|   909K|     1/0/0|         |
  |  10 |           TABLE ACCESS FULL                   | DIM_PERFORMANCE           |      1 |      5 |      5 |00:00:00.01 |       2 |      0 |      0 |       |       |          |         |
  |  11 |           NESTED LOOPS OUTER                  |                           |      1 |    179 |   8631 |00:02:59.06 |   98917 |  40067 |      0 |       |       |          |         |
  |  12 |            NESTED LOOPS                       |                           |      1 |    179 |   8631 |00:02:58.87 |   98917 |  40067 |      0 |       |       |          |         |
  |* 13 |             HASH JOIN                         |                           |      1 |    191 |   8758 |00:00:25.47 |   18101 |   2398 |      0 |  1000K|  1000K|     1/0/0|         |
  |  14 |              TABLE ACCESS FULL                | DIM_TYP_FLUIDE            |      1 |      3 |      3 |00:00:00.01 |       2 |      0 |      0 |       |       |          |         |
  |* 15 |              HASH JOIN                        |                           |      1 |    191 |   8758 |00:00:25.32 |   18099 |   2398 |      0 |   955K|   955K|     1/0/0|         |
  |  16 |               TABLE ACCESS FULL               | DIM_TYP_PDC               |      1 |      3 |      3 |00:00:00.01 |       2 |      0 |      0 |       |       |          |         |
  |  17 |               NESTED LOOPS                    |                           |      1 |        |   8758 |00:00:25.19 |   18097 |   2398 |      0 |       |       |          |         |
  |  18 |                NESTED LOOPS                   |                           |      1 |    191 |   8758 |00:00:06.46 |    9339 |    315 |      0 |       |       |          |         |
  |* 19 |                 HASH JOIN                     |                           |      1 |    191 |   8758 |00:00:00.67 |     565 |      0 |      0 |   880K|   880K|     1/0/0|         |
  |  20 |                  TABLE ACCESS FULL            | DIM_OFFRE                 |      1 |     22 |     22 |00:00:00.01 |       2 |      0 |      0 |       |       |          |         |
  |* 21 |                  HASH JOIN                    |                           |      1 |    191 |   8758 |00:00:00.53 |     563 |      0 |      0 |   779K|   779K|     1/0/0|         |
  |  22 |                   NESTED LOOPS                |                           |      1 |        |      1 |00:00:00.01 |       5 |      0 |      0 |       |       |          |         |
  |  23 |                    NESTED LOOPS               |                           |      1 |      1 |      1 |00:00:00.01 |       4 |      0 |      0 |       |       |          |         |
  |* 24 |                     TABLE ACCESS FULL         | DIM_CONTRAT               |      1 |      1 |      1 |00:00:00.01 |       3 |      0 |      0 |       |       |          |         |
  |* 25 |                     INDEX UNIQUE SCAN         | PK_CONTRACTANT_KEY        |      1 |      1 |      1 |00:00:00.01 |       1 |      0 |      0 |       |       |          |         |
  |* 26 |                    TABLE ACCESS BY INDEX ROWID| DIM_CONTRACTANT           |      1 |      1 |      1 |00:00:00.01 |       1 |      0 |      0 |       |       |          |         |
  |  27 |                   TABLE ACCESS FULL           | REL_SERVICE_POINT_CONTRAT |      1 |  54923 |  54923 |00:00:00.14 |     558 |      0 |      0 |       |       |          |         |
  |* 28 |                 INDEX UNIQUE SCAN             | PK_SP_KEY                 |   8758 |      1 |   8758 |00:00:05.66 |    8774 |    315 |      0 |       |       |          |         |
  |  29 |                TABLE ACCESS BY INDEX ROWID    | DIM_SERVICE_POINT         |   8758 |      1 |   8758 |00:00:18.58 |    8758 |   2083 |      0 |       |       |          |         |
  |  30 |             PARTITION RANGE SINGLE            |                           |   8758 |      1 |   8631 |00:02:33.32 |   80816 |  37669 |      0 |       |       |          |         |
  |* 31 |              TABLE ACCESS BY LOCAL INDEX ROWID| FAC_DAILY_TRANS_PERFORM   |   8758 |      1 |   8631 |00:02:33.18 |   80816 |  37669 |      0 |       |       |          |         |
  |* 32 |               INDEX RANGE SCAN                | IDX3_FAC_DAY_TRAN_PERF    |   8758 |      6 |    103K|00:00:16.05 |   17929 |   4751 |      0 |       |       |          |         |
  |  33 |            TABLE ACCESS BY INDEX ROWID        | DIM_DIAGNOSYS             |   8631 |      1 |      0 |00:00:00.12 |       0 |      0 |      0 |       |       |          |         |
  |* 34 |             INDEX UNIQUE SCAN                 | PK_DIAGNOSYS_KEY          |   8631 |      1 |      0 |00:00:00.05 |       0 |      0 |      0 |       |       |          |         |
  |  35 |          TABLE ACCESS FULL                    | DIM_DATE                  |      1 |  21914 |  21914 |00:00:00.05 |     551 |     28 |      0 |       |       |          |         |
  |  36 |         TABLE ACCESS BY INDEX ROWID           | DIM_RADIO_MODULE          |   8631 |      1 |   8631 |00:00:32.65 |   25895 |   6330 |      0 |       |       |          |         |
  |* 37 |          INDEX UNIQUE SCAN                    | PK_RADMODKEY              |   8631 |      1 |   8631 |00:00:08.38 |   17264 |   2122 |      0 |       |       |          |         |
  |  38 |        TABLE ACCESS BY INDEX ROWID            | DIM_METER_MODULE          |   8631 |      1 |   8631 |00:00:31.51 |   25895 |   6503 |      0 |       |       |          |         |
  |* 39 |         INDEX UNIQUE SCAN                     | PK_METMODKEY              |   8631 |      1 |   8631 |00:00:10.17 |   17264 |   2212 |      0 |       |       |          |         |
  |  40 |       TABLE ACCESS BY INDEX ROWID             | DIM_CUSTOMER              |   8631 |      1 |   8631 |00:01:09.66 |   25895 |   8468 |      0 |       |       |          |         |
  |* 41 |        INDEX UNIQUE SCAN                      | PK_CUSTOMER               |   8631 |      1 |   8631 |00:00:12.03 |   17264 |   2328 |      0 |       |       |          |         |
  |* 42 |      INDEX UNIQUE SCAN                        | PK_METER                  |   8631 |      1 |   8631 |00:00:09.63 |   17264 |   2372 |      0 |       |       |          |         |
  |  43 |     TABLE ACCESS BY INDEX ROWID               | DIM_METER                 |   8631 |      1 |   8631 |00:00:30.71 |    8631 |   5369 |      0 |       |       |          |         |
  --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
   
  Predicate Information (identified by operation id):
  ---------------------------------------------------
   
     8 - access("T786"."DATE_KEY"="T1830"."DATE_KEY")
     9 - access("T786"."PERFORMANCE_KEY"="T2158"."PERFORMANCE_KEY")
    13 - access("T160"."FLUIDE_KEY"="T210"."FLUIDE_KEY")
    15 - access("T160"."TYP_PDC_KEY"="T217"."TYP_PDC_KEY")
    19 - access("T122"."OFFRE_KEY"="T298"."OFFRE_KEY")
    21 - access("T24"."CONTRAT_KEY"="T298"."CONTRAT_KEY" AND "T17"."CONTRACTANT_KEY"="T298"."CONTRACTANT_KEY")
    24 - filter("T24"."NOM_CONTRAT"='COMMUNAY ET REGION (SIE)')
    25 - access("T17"."CONTRACTANT_KEY"="T24"."CONTRACTANT_KEY")
    26 - filter("T17"."NOM_CONTRACTANT"='ER Rhône Alpes Auvergne')
    28 - access("T160"."SERVICE_POINT_KEY"="T298"."SERVICE_POINT_KEY")
    31 - filter("T786"."REPORTING_DATE"=TO_DATE(' 2013-02-09 00:00:00', 'syyyy-mm-dd hh24:mi:ss'))
    32 - access("T160"."SERVICE_POINT_KEY"="T786"."SERVICE_POINT_KEY")
    34 - access("T786"."DIAGNOSYS_KEY"="T2186"."DIAGNOSYS_KEY")
    37 - access("T786"."RADIO_MODULE_KEY"="T949"."RADIO_MODULE_KEY")
    39 - access("T786"."METER_MODULE_KEY"="T1191"."METER_MODULE_KEY")
    41 - access("T786"."CUSTOMER_KEY"="T2098"."CUSTOMER_KEY")
    42 - access("T786"."METER_KEY"="T1011"."METER_KEY")
   

  Also, line 31, my score is used with a local index and E-lines and A-lines are correct, but the index takes 02:33 minutes to get a 8758 ranks who seems to be excessive. EC which could I check/do to reduce it?
  

  |  30 |             PARTITION RANGE SINGLE            |                           |   8758 |      1 |   8631 |00:02:33.32 |   80816 |  37669 |      0 |       |       |          |         |
  |* 31 |              TABLE ACCESS BY LOCAL INDEX ROWID| FAC_DAILY_TRANS_PERFORM   |   8758 |      1 |   8631 |00:02:33.18 |   80816 |  37669 |      0 |       |       |          |         |

  Thank you very much!

  Hello

  (1) for a significant query, it is rare that the optimizer obtains all right cardinalities
  (2) but normally, it is not really necessary, in most cases just to be in the right ballpark (1-2 orders of magnitude)
  (3) as others already said, the optimizer in your case does that 1 error - in step 1, when evaluating a cardinality of a hash of several columns, the steps join upward simply reproduce this error over and over
  (4) rethink the cardinality of the join is always difficult, and especially when it comes to columns multiple joins because you have several possibilities made a mistake in the selectivity of unique columns, but also because there is a chance that there is a correlation between the columns attached; also, the optimizer uses many settings, called 'mental health controls', which does not make things more simple
  (5) you can try to create statistics on a column in the joined columns group for a more accurate forecast of join cardinality
  (6) there again, get all the cardinalities right does not guarantee that you will get a good plan - for example, you may be limited by the lack of an appropriate index
  (7) sometimes, and more feedback of cardinality (or even instead) it is better to tackle the problem from a different angle: see what operations are ineffective (e.g. rejection lines after spending a lot of e/s on their acquisition), see http://savvinov.com/2013/01/28/efficiency-based-sql-tuning/
  (8) in your case, you have noticed that you: step

  |* 31 |              TABLE ACCESS BY LOCAL INDEX ROWID| FAC_DAILY_TRANS_PERFORM   |   8758 |      1 |   8631 |00:02:33.18 |   80816 |  37669 |      0 |       |       |          |  
  

  is the biggest problem for this plan, even if the cardinality estimate of it's very precise
  (9) you can fix the problem by extending the IDX3_FAC_DAY_TRAN_PERF index to the REPORTING_DATE column: If you do this, then you won't need to 63 k logical reads to return lines k 103 just to reject 90% of them - you will reject them at an earlier stage (INDEX RANGE SCAN) and only need to at most 8 k readings to acquire the filtered lines by table rowid

  Best regards
  Nikolai