Which execution is more performance effective in SQL Query

Hi
Any one help me regarding oracle performance, which is best for application performance
1) I have multiple option and store in application and the option value or text store in transaction table, and we are querying information filter by option field value.
2) I have a table stored all option with a primary key and select the option query by table , and the value stored in transaction table as foregin key, we qre querying information with joining of two table and filter by option table column
in this scenario , i have two of more question
1) which is faster process for information display
2) which option will use minimum memory
3) which option will effect application performance
4) which option will effect database performance
which is best for application performation ?
Thanks

Hi
I am not a student,
I am analysis the execution time of those query as below
1) Select Name , Age, Country,Zone from Countrydata where zone='EAST'
2) Select a.Name, a.Age, a.Country,b.zone from countrydata a, countryzone b where a.zonecode=b.zonecode and a.zonecode=1
both query result are same
query case 1- No index
query case -2 with index
now please suggest me which as per below my query
1) which query execution time will be short
2) in which query data storage increase or dcrease in the table and data size and what will impact on database performance
3) which query is best for application developer
4) which is best as per data modeling
Regards

Similar Messages

  • Need help in improving the performance for the sql query

    Thanks in advance for helping me.
    I was trying to improve the performance of the below query. I tried the following methods used merge instead of update, used bulk collect / Forall update, used ordered hint, created a temp table and upadated the target table using the same. The methods which I used did not improve any performance. The data count which is updated in the target table is 2 million records and the target table has 15 million records.
    Any suggestions or solutions for improving performance are appreciated
    SQL query:
    update targettable tt
    set mnop = 'G',
    where ( x,y,z ) in
    select a.x, a.y,a.z
    from table1 a
    where (a.x, a.y,a.z) not in (
    select b.x,b.y,b.z
    from table2 b
    where 'O' = b.defg
    and mnop = 'P'
    and hijkl = 'UVW';

    987981 wrote:
    I was trying to improve the performance of the below query. I tried the following methods used merge instead of update, used bulk collect / Forall update, used ordered hint, created a temp table and upadated the target table using the same. The methods which I used did not improve any performance. And that meant what? Surely if you spend all that time and effort to try various approaches, it should mean something? Failures are as important teachers as successes. You need to learn from failures too. :-)
    The data count which is updated in the target table is 2 million records and the target table has 15 million records.Tables have rows btw, not records. Database people tend to get upset when rows are called records, as records exist in files and a database is not a mere collection of records and files.
    The failure to find a single faster method with the approaches you tried, points to that you do not know what the actual performance problem is. And without knowing the problem, you still went ahead, guns blazing.
    The very first step in dealing with any software engineering problem, is to identify the problem. Seeing the symptoms (slow performance) is still a long way from problem identification.
    Part of identifying the performance problem, is understanding the workload. Just what does the code task the database to do?
    From your comments, it needs to find 2 million rows from 15 million rows. Change these rows. And then write 2 million rows back to disk.
    That is not a small workload. Simple example. Let's say that the 2 million row find is 1ms/row and the 2 million row write is also 1ms/row. This means a 66 minute workload. Due to the number of rows, an increase in time/row either way, will potentially have 2 million fold impact.
    So where is the performance problem? Time spend finding the 2 million rows (where other tables need to be read, indexes used, etc)? Time spend writing the 2 million rows (where triggers and indexes need to be fired and maintained)? Both?

  • Performance of native sql query detoriates

    Dear Experts,
    The performance of my native SQL query is bad. On the database the query takes less than 5 seconds to process. From my abap program I get a session timeout dump after 10 minutes. What might be the possible reason.
    Warm Regards,
    Abdullah

    I am not a DBA, but this is a wild guess.
    I have a native SQL query. It was running fine all morning(transported it to production today). By afternoon the report was not giving any output.
    I went to the MS SQL query analyzer and executed the query, it returned the results in less than 5 seconds. The same query when I was executing from SAP using native SQL took more than 10 minutes and gave a dump(time exceeded).
    My database guy asked me to execute the following on the database. Dbcc dbreindex('tablename')
    The report is running fine since then. I am still not satisfied if this is the reason the performance is back on track, but yeah the report is running fine again. There seems to be some problem with the indexes.
    I am using standard classes provided by SAP to execute my query and after execution the resultset reference object is being closed, I am closing the connection.
    the code is as below.
          PERFORM:
            connect               USING con_name con_ref,
            select_into_table     USING con_ref,
            disconnect            USING con_ref.
    *  FORM connect
    *  Connects to the database specified by the logical connection name
    *  P_CON_NAME which is expected to be specified in table DBCON. In case
    *  of success the form returns in P_CON_REF a reference to a connection
    *  object of class CL_SQL_CONNECTION.
    *  --> P_CON_NAME  logical connection name
    *  <-- P_CON_REF   reference to a CL_SQL_CONNECTION object
    FORM connect  USING    p_con_name TYPE dbcon-con_name
                           p_con_ref  TYPE REF TO cl_sql_connection
                           RAISING cx_sql_exception.
    * if CON_NAME is not initial then try to open the connection, otherwise
    * create a connection object representing the default connection.
      IF p_con_name IS INITIAL.
        CREATE OBJECT p_con_ref.
      ELSE.
        p_con_ref = cl_sql_connection=>get_connection( p_con_name ).
      ENDIF.
    ENDFORM.                    " connect
    *  FORM select_into_table
    *  Selects some rows from the test table and fetches the result rows
    *  into an internal table whose row structure corresponds to the
    *  queries select list columns.
    FORM select_into_table
      USING   p_con_ref TYPE REF TO cl_sql_connection
      RAISING cx_sql_exception.
      DATA:
        l_stmt         TYPE string,
        l_stmt_ref     TYPE REF TO cl_sql_statement,
        l_dref         TYPE REF TO data,
        l_res_ref      TYPE REF TO cl_sql_result_set,
    *Data related query
        l_itab         TYPE TABLE OF t_pricing_report,
        l_row_cnt      TYPE i.
    * create the query string
    CONCATENATE
        'select A.SEQ,A.CONDTABLE,A.CONDNAME,A.VKORG,A.VTWEG,A.MATKL,A.MATNR,B.MTEXT,A.VKGRP,A.SGRPNAME,'
        'A.VKBUR,A.SOFFNAME,A.ZSALES,A.SCNTNAME,A.KUNNR,A.SCSTNAME,A.PRBATCH,A.INCO1,'
        'A.INCO2,A.DATAB,A.DATBI,A.KBETR,A.KONWA,A.KOSRT,B.MTART,B.GROES,B.VOLUM,B.EXTWG,B.WRKST,'
        'A.MXWRT,A.GKWRT,'
        'B.PATTERN,B.RIM,B.SERIES,B.SPDINDEX,B.LDINDX,B.MGROUP,B.APPLN,B.SDWALL,B.MGRPTXT'
        'FROM Z_PRICELIST A,Z_MATERIALVIEW B'
        'WHERE A.MANDT = ? AND'
              'B.MANDT = A.MANDT AND'
              'A.MATNR = B.MATNR AND'
              'A.KSCHL = ? AND'
              'A.CONDTABLE LIKE ? AND'
              'A.VKORG LIKE ? AND'
              'A.VTWEG LIKE ? AND'
              'A.MATKL >= ? AND A.MATKL <= ? AND'
              'A.MATNR >= ? AND A.MATNR <= ? AND'
              'A.INCO1 LIKE ? AND'
              'A.INCO2 LIKE ? AND'
              'A.ZSALES >= ? AND A.ZSALES <= ? AND'
              'A.KUNNR  >= ? AND A.KUNNR <= ? AND'
              'A.PRBATCH  >= ? AND A.PRBATCH <= ? AND'
              'A.VKBUR  >= ? AND A.VKBUR <= ? AND'
              'A.VKGRP  >= ? AND A.VKGRP <= ? AND'
              'B.WRKST  >= ? AND B.WRKST <= ? AND'
              'B.MTART  >= ? AND B.MTART <= ? AND'
              '? BETWEEN A.DATAB AND A.DATBI AND'
              'B.GROES LIKE ? AND'
              'B.LDINDX LIKE ? AND'
              'B.SPDINDEX LIKE ? AND'
              'B.RIM LIKE ? AND'
              'B.SERIES LIKE ? AND'
              'B.PATTERN LIKE ? AND'
              'B.MGROUP LIKE ?'
              'order by A.MATNR'
        INTO l_stmt SEPARATED BY space.                         "#EC NOTEXT
    * create a statement object
      l_stmt_ref = p_con_ref->create_statement( ).
    * bind input variables
      GET REFERENCE OF l_col1 INTO l_dref.
      l_stmt_ref->set_param( l_dref ).
    *binding other references here
      GET REFERENCE OF l_col33 INTO l_dref.
      l_stmt_ref->set_param( l_dref ).
    * set the input values and execute the query
      l_col1  = sy-mandt.
    *..Assigning values here
      l_col33 = p_mgroup.
    *  PERFORM trace_2 USING 'EXECUTE_QUERY' l_stmt l_col1 l_col2.
      l_res_ref = l_stmt_ref->execute_query( l_stmt ).
    * set output table
      GET REFERENCE OF l_itab INTO l_dref.
      l_res_ref->set_param_table( l_dref ).
    * get the complete result set
      l_row_cnt = l_res_ref->next_package( ).
    * display the contents of the output table
    *  PERFORM trace_next_package USING l_itab.
    *  PERFORM trace_result USING l_row_cnt 'rows fetched'.
      pricing_report[] = l_itab[].
      free l_itab.
    * don't forget to close the result set object in order to free
    * resources on the database
      l_res_ref->close( ).
    ENDFORM.                    "select_into_table
    *  FORM disconnect
    *  Disconnect from the given connection. In case of the default
    *  connection this can be omitted.
    FORM disconnect
      USING   p_con_ref TYPE REF TO cl_sql_connection
      RAISING cx_sql_exception.
      DATA: l_con_name TYPE dbcon-con_name.
      l_con_name = p_con_ref->get_con_name( ).
      CHECK l_con_name <> cl_sql_connection=>c_default_connection.
    *  PERFORM trace_0 USING 'CLOSE CONNECTION' l_con_name.
      p_con_ref->close( ).
    *  PERFORM trace_result USING l_con_name 'closed'.
    ENDFORM.                    "disconnect
    *  FORM handle_sql_exception
    *  Write appropriate error messages when a SQL exception has occured
    *  -->  P_SQLERR_REF  reference to a CX_SQL_EXCEPTION object
    FORM handle_sql_exception
      USING p_sqlerr_ref TYPE REF TO cx_sql_exception.
      FORMAT COLOR COL_NEGATIVE.
      IF p_sqlerr_ref->db_error = 'X'.
        WRITE: / 'SQL error occured:', p_sqlerr_ref->sql_code,
               / p_sqlerr_ref->sql_message.                     "#EC NOTEXT
      ELSE.
        WRITE:
          / 'Error from DBI (details in dev-trace):',
            p_sqlerr_ref->internal_error.                       "#EC NOTEXT
      ENDIF.
    ENDFORM.                    "handle_sql_exception

  • Performance issue with sql query. Please explain

    I have a sql query
    A.column1 and B.column1 are indexed.
    Query1: select A.column1,A.column3, B.column2 from tableA , tableB where A.column1=B.column1;
    query2: select A.column1,A.column3,B.column2,B.column4 from tableA , tableB where A.column1=B.column1;
    1. Does both query takes same time? If not why?.
    As they both go for same row with different number of columns. And Since the complete datablock is loaded in Database buffer cache. so there should not be extra time taken upto this time.
    Please tell me if I am wrong.

    For me apart from required excessive bytes sent via SQL*Net to client as well bytes received via SQL*Net from client will cause to chatty network which will degrade the performance.
    SQL> COLUMN plan_plus_exp FORMAT A100
    SQL> SET LINESIZE 1000
    SQL> SET AUTOTRACE TRACEONLY
    SQL> SELECT *
      2    FROM emp
      3  /
    14 rows selected.
    Execution Plan
       0      SELECT STATEMENT Optimizer=ALL_ROWS (Cost=3 Card=14 Bytes=616)
       1    0   TABLE ACCESS (FULL) OF 'EMP' (TABLE) (Cost=3 Card=14 Bytes=616)
    Statistics
              1  recursive calls
              0  db block gets
              8  consistent gets
              0  physical reads
              0  redo size
    1631 bytes sent via SQL*Net to client
    423 bytes received via SQL*Net from client
              2  SQL*Net roundtrips to/from client
              0  sorts (memory)
              0  sorts (disk)
             14  rows processed
    SQL> SELECT ename
      2    FROM emp
      3  /
    14 rows selected.
    Execution Plan
       0      SELECT STATEMENT Optimizer=ALL_ROWS (Cost=3 Card=14 Bytes=154)
       1    0   TABLE ACCESS (FULL) OF 'EMP' (TABLE) (Cost=3 Card=14 Bytes=154)
    Statistics
              1  recursive calls
              0  db block gets
              8  consistent gets
              0  physical reads
              0  redo size
    456 bytes sent via SQL*Net to client
    423 bytes received via SQL*Net from client
              2  SQL*Net roundtrips to/from client
              0  sorts (memory)
              0  sorts (disk)
             14  rows processed Khurram

  • Performance Problem with SQL Query

    Hi
    I have a SQL Query (say, SSS) which runs quite fast when I run it from TOAD.
    But when the same query is being used as
    INSERT INTO <table> VALUES <SSS>;
    It is taking hours to complete.
    The Original Query(OSQ) was changed to this SQL query(SSS) for Performance improvement.
    The Cost has also improved quite a lot from OSQ to SSS. But when SSS runs from inside a procedure it is taking the same long hours as OSQ used to take.
    Please help ASAP as it needs to get fixed.
    Thanks
    Arnab

    SELECT SRM.ID PROJECT_ID,
    SRM.UNIQUE_NAME PROJECT_CODE,
    ODFP.GS_FINANCE_PROJ_CODE,
    ODFP.GS_PRODUCT_CODE,
    ODFP.GS_CUSTOMER,
    ODFP.GS_LEGAL_ENT_REF,
    ODF_PRJ_TYPE.NAME GS_PROJ_TYPE,
    ODF_SRC_SYS.NAME GS_SOURCE_SYSTEM,
    RESM.FIRST_NAME||' '||RESM.LAST_NAME GS_PROJECT_MANAGER,
    ODFP.GS_LEG_PROJ_NUM,
    ODFP.GS_SUPP_DOC ,
    ODFP.GS_REQUEST_ID ,
    ODFP.GS_CONTRACT ,
    ODFP.GS_SIEBEL_REF,
    DECODE(budg.ODF_PARENT_ID,NULL,PRJP.bdgt_cst_total, budg.lab_budg)LAB_BDGT,
    DECODE(budg.ODF_PARENT_ID,NULL,0,budg.nli_budg)NLI_BDGT,
    (DECODE(budg.ODF_PARENT_ID,NULL,PRJP.bdgt_cst_total, budg.lab_budg))+(DECODE(budg.ODF_PARENT_ID,NULL,0,budg.nli_budg)) sum_lab_nonlab_budg,
    frct.LABOUR,
    frct.NON_LABOUR,
    frct.LABOUR+frct.NON_LABOUR FORCAST_TOT_SUM,
    gfr.code GFR_CODE,
    odfp.gs_rev_recognition,
    rev_rec.NAME gs_rev_recognition,
    DECODE(odfp.gs_prevent_reqtxn,1,'YES','0','NO','NO'),
    GS_PROGTYPE.NAME,
    COMM_APPR.UNIQUE_NAME COMM_APPROVER_EIN,
    COMM_APPR.FIRST_NAME||' '||COMM_APPR.LAST_NAME COMM_APPROVER_NAME,
    PR_APPR.UNIQUE_NAME PR_APPROVER_EIN,
    PR_APPR.FIRST_NAME||' '||PR_APPR.LAST_NAME PR_APPROVER_NAME,
    ODFP.GS_BILL_BUD,
    frct.FRCST_REVENUE TOTAL_FRC_REVENUE,
    actuals.LABOUR_ACTUALS,
    actuals.NL_ACTUALS,
    -- get_wip_totalcost(prjp.prid,'L') LABOUR_ACTUALS,
    -- get_wip_totalcost(prjp.prid,'M') NL_ACTUALS,
    ODFP.GS_ASS_PRODUCT_CODE,
    PROJ_TEMPLATE.ID GS_PROJ_TEMPLATE_ID,
    PROJ_TEMPLATE.UNIQUE_NAME GS_PROJ_TEMPLATE_CODE,
    PROJ_TEMPLATE.name GS_PROJ_TEMPLATE_NAME,
    DECODE( NVL(ODFP.GS_FIN_TEMPLATE,0) ,0,'NO',1,'YES') GS_FIN_TEMPLATE,
    ODF_PRJ_TYPE.LOOKUP_CODE GS_PROJ_TYPE
    FROM PROJECTS SRM,
    J_PROJECTS PRJP,
    A_PROJECT ODFP,
    RESOURCES RESM,
    s_gfr gfr,
    RESOURCES PROJ_ACCT,
    RESOURCES COMM_APPR,
    RESOURCES PR_APPR,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_PROJECT_TYPE'
    AND LOOKUP.ID =CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS') ODF_PRJ_TYPE,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_SOURCE_SYSTEM'
    AND LOOKUP.ID =CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS') ODF_SRC_SYS,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_PROG_TYPE'
    AND LOOKUP.ID =CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS')GS_PROGTYPE,
    (SELECT odf_parent_id
    ,SUM(gs_lab_budg) lab_budg
    ,SUM(gs_nl_budg) nli_budg
    FROM T_CHANGES
    WHERE gs_status='APPR'
    GROUP BY odf_parent_id) budg,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_REV_RECOGNITION'
    AND LOOKUP.ID = CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS') rev_rec,
    (SELECT FRCP1.project_id,
    NVL(SUM(CASE WHEN srmr.resource_type=0 THEN for_val.cost ELSE NULL END),0) AS LABOUR,
    NVL(SUM(CASE WHEN srmr.resource_type <> 0 THEN for_val.cost ELSE NULL END),0) AS NON_LABOUR,
    NVL(SUM(for_val.revenue),0) AS FRCST_REVENUE
    FROM T_PROPERTIES FRCP1,
    T_DETAILS FOR_DET,
    RESOURCES SRMR,
    T_VALUES FOR_VAL
    WHERE (FRCP1.project_id, revision) IN (SELECT project_id, MAX(revision)
    FROM T_PROPERTIES FRCP
    WHERE FRCP.status=2
    AND FRCP.PERIOD_TYPE='SEMI_MONTHLY'
    GROUP BY project_id)
    AND FRCP1.PERIOD_TYPE='SEMI_MONTHLY'
    AND FOR_DET.forecast_id=frcp1.id
    AND SRMR.id(+)=FOR_DET.detail_id
    AND FOR_VAL.currency_type='BILLING'
    AND FOR_VAL.forecast_details_id=FOR_DET.ID
    GROUP BY FRCP1.project_id) frct,
    (SELECT srmp.id
    ,NVL(SUM(CASE WHEN wip.transtype='L' THEN WIPVAL.TOTALCOST ELSE NULL END),0) AS LABOUR_ACTUALS
    ,NVL(SUM(CASE WHEN wip.transtype='M' THEN WIPVAL.TOTALCOST ELSE NULL END),0) AS NL_ACTUALS
    FROM A_WIP WIP
    ,om_periods biz
    ,P_VALUES WIPVAL
    ,PROJECTS SRMP
    ,A_PROJECT ODF
    WHERE biz.period_type ='SEMI_MONTHLY'
    AND TRUNC(wip.TRANSDATE) BETWEEN biz.start_date AND biz.end_date
    AND WIP.TRANSNO=WIPVAL.TRANSNO
    AND WIPVAL.CURRENCY_TYPE='BILLING'
    AND SRMP.UNIQUE_NAME=WIP.PROJECT_CODE
    AND SRMP.ID=ODF.ID
    AND UPPER(ODF.partition_code)='GLOBAL'
    AND UPPER(WIP.external_id) <> 'SPREADSHEET'
    GROUP BY srmp.id) actuals,
    RESOURCES BUD_HOLDER,
    PROJECTS PROJ_TEMPLATE
    WHERE SRM.ID = PRJP.PRID
    AND ODFP.ID=PRJP.PRID
    AND ODFP.PARTITION_CODE='GLOBAL'
    AND RESM.id(+)=prjp.manager_id
    AND ODFP.GS_PROJECT_TYPE = ODF_PRJ_TYPE.LOOKUP_CODE(+)
    AND ODFP.GS_SOURCE_SYSTEM = ODF_SRC_SYS.LOOKUP_CODE(+)
    AND ODFP.GS_prog_type = GS_PROGTYPE.LOOKUP_CODE(+)
    AND budg.odf_parent_id(+)=SRM.ID
    AND frct.project_id(+) = prjp.prid
    AND ODFP.gs_risk_gfr=gfr.code(+)
    AND ODFP.GS_REV_RECOGNITION=rev_rec.LOOKUP_CODE(+)
    AND ODFP.GS_PROJ_ACCT = PROJ_ACCT.ID(+)
    AND ODFP.GS_COMM_APPR = COMM_APPR.ID(+)
    AND ODFP.GS_PR_APPR = PR_APPR.ID(+)
    AND actuals.id(+) = prjp.prid -- Arnab
    AND ODFP.GS_BUDGET_HOLDER=BUD_HOLDER.ID(+)
    AND ODFP.GS_TEMPLATE_USED=PROJ_TEMPLATE.ID(+)
    The above is OSQ with cost 45000. ... takes 1-1.5 hours through Toad
    SELECT SRM.ID PROJECT_ID,
    SRM.UNIQUE_NAME PROJECT_CODE,
    ODFP.GS_FINANCE_PROJ_CODE,
    ODFP.GS_PRODUCT_CODE,
    ODFP.GS_CUSTOMER,
    ODFP.GS_LEGAL_ENT_REF,
    ODF_PRJ_TYPE.NAME GS_PROJ_TYPE,
    ODF_SRC_SYS.NAME GS_SOURCE_SYSTEM,
    RESM.FIRST_NAME||' '||RESM.LAST_NAME GS_PROJECT_MANAGER,
    ODFP.GS_LEG_PROJ_NUM,
    ODFP.GS_SUPP_DOC ,
    ODFP.GS_REQUEST_ID ,
    ODFP.GS_CONTRACT ,
    ODFP.GS_SIEBEL_REF,
    DECODE(budg.ODF_PARENT_ID,NULL,PRJP.bdgt_cst_total, budg.lab_budg)LAB_BDGT,
    DECODE(budg.ODF_PARENT_ID,NULL,0,budg.nli_budg)NLI_BDGT,
    (DECODE(budg.ODF_PARENT_ID,NULL,PRJP.bdgt_cst_total, budg.lab_budg))+(DECODE(budg.ODF_PARENT_ID,NULL,0,budg.nli_budg)) sum_lab_nonlab_budg,
    frct.LABOUR,
    frct.NON_LABOUR,
    frct.LABOUR+frct.NON_LABOUR FORCAST_TOT_SUM,
    gfr.code GFR_CODE,
    odfp.gs_rev_recognition,
    rev_rec.NAME gs_rev_recognition,
    DECODE(odfp.gs_prevent_reqtxn,1,'YES','0','NO','NO'),
    GS_PROGTYPE.NAME,
    COMM_APPR.UNIQUE_NAME COMM_APPROVER_EIN,
    COMM_APPR.FIRST_NAME||' '||COMM_APPR.LAST_NAME COMM_APPROVER_NAME,
    PR_APPR.UNIQUE_NAME PR_APPROVER_EIN,
    PR_APPR.FIRST_NAME||' '||PR_APPR.LAST_NAME PR_APPROVER_NAME,
    ODFP.GS_BILL_BUD,
    frct.FRCST_REVENUE TOTAL_FRC_REVENUE,
    -- actuals.LABOUR_ACTUALS, -- Arnab
    -- actuals.NL_ACTUALS, -- Arnab
    get_wip_totalcost(prjp.prid,'L') LABOUR_ACTUALS, -- Arnab
    get_wip_totalcost(prjp.prid,'M') NL_ACTUALS, -- Arnab
    ODFP.GS_ASS_PRODUCT_CODE,
    PROJ_TEMPLATE.ID GS_PROJ_TEMPLATE_ID,
    PROJ_TEMPLATE.UNIQUE_NAME GS_PROJ_TEMPLATE_CODE,
    PROJ_TEMPLATE.name GS_PROJ_TEMPLATE_NAME,
    DECODE( NVL(ODFP.GS_FIN_TEMPLATE,0) ,0,'NO',1,'YES') GS_FIN_TEMPLATE,
    ODF_PRJ_TYPE.LOOKUP_CODE GS_PROJ_TYPE
    FROM PROJECTS SRM,
    J_PROJECTS PRJP,
    A_PROJECT ODFP,
    RESOURCES RESM,
    s_gfr gfr,
    RESOURCES PROJ_ACCT,
    RESOURCES COMM_APPR,
    RESOURCES PR_APPR,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_PROJECT_TYPE'
    AND LOOKUP.ID =CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS') ODF_PRJ_TYPE,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_SOURCE_SYSTEM'
    AND LOOKUP.ID =CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS') ODF_SRC_SYS,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_PROG_TYPE'
    AND LOOKUP.ID =CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS')GS_PROGTYPE,
    (SELECT odf_parent_id
    ,SUM(gs_lab_budg) lab_budg
    ,SUM(gs_nl_budg) nli_budg
    FROM T_CHANGES
    WHERE gs_status='APPR'
    GROUP BY odf_parent_id) budg,
    (SELECT CAP.NAME
    ,LOOKUP.LOOKUP_CODE
    FROM S_NLS CAP,
    N_LOOKUPS LOOKUP
    WHERE LOOKUP_TYPE ='GS_REV_RECOGNITION'
    AND LOOKUP.ID = CAP.PK_ID
    AND CAP.LANGUAGE_CODE='en'
    AND CAP.TABLE_NAME = 'CMN_LOOKUPS') rev_rec,
    (SELECT FRCP1.project_id,
    NVL(SUM(CASE WHEN srmr.resource_type=0 THEN for_val.cost ELSE NULL END),0) AS LABOUR,
    NVL(SUM(CASE WHEN srmr.resource_type <> 0 THEN for_val.cost ELSE NULL END),0) AS NON_LABOUR,
    NVL(SUM(for_val.revenue),0) AS FRCST_REVENUE
    FROM T_PROPERTIES FRCP1,
    T_DETAILS FOR_DET,
    RESOURCES SRMR,
    T_VALUES FOR_VAL
    WHERE (FRCP1.project_id, revision) IN (SELECT project_id, MAX(revision)
    FROM T_PROPERTIES FRCP
    WHERE FRCP.status=2
    AND FRCP.PERIOD_TYPE='SEMI_MONTHLY'
    GROUP BY project_id)
    AND FRCP1.PERIOD_TYPE='SEMI_MONTHLY'
    AND FOR_DET.forecast_id=frcp1.id
    AND SRMR.id(+)=FOR_DET.detail_id
    AND FOR_VAL.currency_type='BILLING'
    AND FOR_VAL.forecast_details_id=FOR_DET.ID
    GROUP BY FRCP1.project_id) frct,
    /* (SELECT srmp.id
    ,NVL(SUM(CASE WHEN wip.transtype='L' THEN WIPVAL.TOTALCOST ELSE NULL END),0) AS LABOUR_ACTUALS
    ,NVL(SUM(CASE WHEN wip.transtype='M' THEN WIPVAL.TOTALCOST ELSE NULL END),0) AS NL_ACTUALS
    FROM A_WIP WIP
    ,om_periods biz
    ,P_VALUES WIPVAL
    ,PROJECTS SRMP
    ,A_PROJECT ODF
    WHERE biz.period_type ='SEMI_MONTHLY'
    AND TRUNC(wip.TRANSDATE) BETWEEN biz.start_date AND biz.end_date
    AND WIP.TRANSNO=WIPVAL.TRANSNO
    AND WIPVAL.CURRENCY_TYPE='BILLING'
    AND SRMP.UNIQUE_NAME=WIP.PROJECT_CODE
    AND SRMP.ID=ODF.ID
    AND UPPER(ODF.partition_code)='GLOBAL'
    AND UPPER(WIP.external_id) <> 'SPREADSHEET'
    GROUP BY srmp.id) actuals, */ -- Arnab
    RESOURCES BUD_HOLDER,
    PROJECTS PROJ_TEMPLATE
    WHERE SRM.ID = PRJP.PRID
    AND ODFP.ID=PRJP.PRID
    AND ODFP.PARTITION_CODE='GLOBAL'
    AND RESM.id(+)=prjp.manager_id
    AND ODFP.GS_PROJECT_TYPE = ODF_PRJ_TYPE.LOOKUP_CODE(+)
    AND ODFP.GS_SOURCE_SYSTEM = ODF_SRC_SYS.LOOKUP_CODE(+)
    AND ODFP.GS_prog_type = GS_PROGTYPE.LOOKUP_CODE(+)
    AND budg.odf_parent_id(+)=SRM.ID
    AND frct.project_id(+) = prjp.prid
    AND ODFP.gs_risk_gfr=gfr.code(+)
    AND ODFP.GS_REV_RECOGNITION=rev_rec.LOOKUP_CODE(+)
    AND ODFP.GS_PROJ_ACCT = PROJ_ACCT.ID(+)
    AND ODFP.GS_COMM_APPR = COMM_APPR.ID(+)
    AND ODFP.GS_PR_APPR = PR_APPR.ID(+)
    AND     actuals.id(+) = prjp.prid  Arnab
    AND ODFP.GS_BUDGET_HOLDER=BUD_HOLDER.ID(+)
    AND ODFP.GS_TEMPLATE_USED=PROJ_TEMPLATE.ID(+)
    This one in SSS where "-- Arnab " are only changes .....
    The cost of this 7000.....finishes in 1 - 1.5 min in Toad
    This is used as INSERT INTO TABLE <Select Query> ... this was taking 1.5 Hours to run.
    It was changed using BULK COLLECT - still its taking around 1.5 hours to run...
    Thanks
    Arnab

  • What Indexes should be created for improve performance of the sql query

    Hello Admins
    One of my user is facing slow performance issue while running the below query. Can someone, please guide me for the same. I want to know, what indexes should be created to improve the performance of this query. Also what else can be done to achieve the same.
    SQL Query:-
    SELECT UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_NUMBER))),
    CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_NUMBER,
    CGSBI_SHIP_DIST_S_EXTRACT.PO_SHIPMENT_NUMBER,
    CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_SHIP_DIST_NUMBER,
    CGSBI_SHIP_DIST_S_EXTRACT.DISTRIBUTION_DATE,
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_SHIP_DIST_LINE_ID))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PROJECT_ID))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.ACCOUNT_DISTRIBUTION_CODE))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.ORACLE_ACCOUNT_NUMBER))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.COMPONENT_CODE))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.TRANSACTION_CURRENCY_CODE))),
    CGSBI_SHIP_DIST_S_EXTRACT.ORDER_QUANTITY, UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.ORDER_UOM))),
    CGSBI_SHIP_DIST_S_EXTRACT.UNIT_PRICE_TRX_CURRENCY,
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.EXPENSE_TYPE_INDICATOR))),
    CGSBI_SHIP_DIST_S_EXTRACT.SOR_ID,
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_ITEM_CODE))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_ITEM_DESC))),
    CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_ITEM_LEAD_TIME,
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.UNSPSC_CODE))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.BUYER_ID))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.REQUESTOR_ID))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.APPROVER_ID))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.SUPPLIER_SITE_ID))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.SUPPLIER_GSL_NUMBER))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.SHIP_TO_LOCATION_CODE))),
    UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.TASK_ID))),
    (LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_RELEASE_ID)))
    FROM
    CGSBI_SHIP_DIST_S_EXTRACT
    WHERE PO_NUMBER IS NOT NULL;
    I generated the explain plan for this query and found the following:-
    Explain Plan:-
    SQL> explain plan for SELECT UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_NUMBER))),
    2 CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_NUMBER,
    3 CGSBI_SHIP_DIST_S_EXTRACT.PO_SHIPMENT_NUMBER,
    4 CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_SHIP_DIST_NUMBER,
    5 CGSBI_SHIP_DIST_S_EXTRACT.DISTRIBUTION_DATE,
    6 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_SHIP_DIST_LINE_ID))),
    7 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PROJECT_ID))),
    8 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.ACCOUNT_DISTRIBUTION_CODE))),
    9 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.ORACLE_ACCOUNT_NUMBER))),
    10 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.COMPONENT_CODE))),
    11 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.TRANSACTION_CURRENCY_CODE))),
    12 CGSBI_SHIP_DIST_S_EXTRACT.ORDER_QUANTITY, UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.ORDER_UOM))),
    13 CGSBI_SHIP_DIST_S_EXTRACT.UNIT_PRICE_TRX_CURRENCY,
    14 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.EXPENSE_TYPE_INDICATOR))),
    15 CGSBI_SHIP_DIST_S_EXTRACT.SOR_ID,
    16 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_ITEM_CODE))),
    17 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_ITEM_DESC))),
    18 CGSBI_SHIP_DIST_S_EXTRACT.PO_LINE_ITEM_LEAD_TIME,
    19 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.UNSPSC_CODE))),
    20 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.BUYER_ID))),
    21 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.REQUESTOR_ID))),
    22 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.APPROVER_ID))),
    23 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.SUPPLIER_SITE_ID))),
    24 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.SUPPLIER_GSL_NUMBER))),
    25 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.SHIP_TO_LOCATION_CODE))),
    26 UPPER(LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.TASK_ID))),
    27 (LTRIM(RTRIM(CGSBI_SHIP_DIST_S_EXTRACT.PO_RELEASE_ID)))
    28 FROM
    29 CGSBI_SHIP_DIST_S_EXTRACT
    30 WHERE PO_NUMBER IS NOT NULL;
    Explained.
    SQL>
    SQL>
    SQL> SELECT * FROM TABLE(dbms_xplan.display);
    PLAN_TABLE_OUTPUT
    Plan hash value: 3891180274
    | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
    | 0 | SELECT STATEMENT | | 77647 | 39M| 2006 (1)| 00:00:25 |
    |* 1 | TABLE ACCESS FULL| CGSBI_SHIP_DIST_S_EXTRACT | 77647 | 39M| 2006 (1)| 00:00:25 |
    Predicate Information (identified by operation id):
    1 - filter("PO_NUMBER" IS NOT NULL)
    13 rows selected.
    SQL>
    SQL>
    Kindly suggest on this...
    Thanks & Regards
    -Naveen Gangil
    Oracle DBA

    Rafi is correct. Since po_number is the filter column, the only chance you have for using an index to access the table is on that column. However, if there are few (or none) rows with null po_number, you will always have FTS. Does the table have a PK ( which probably consists of at least po_number, line_number )? If that is the case po_number could never be null and in which case you are dumping the whole table and no indexing scheme is going to improve the queri's performance. You might, repeat might, see performance improvement if you cleanse the data in the table ( to eliminate the need for UPPER(LTRIM)RTRIM())) ) before querying it so the the data does not have to be massaged before returning it.

  • Execution time, elapsed time  of an sql query

    Can you please tell me how to get the execution time, elapsed time of an sql query

    user8680248 wrote:
    I am running query in the database
    I would like to know how long the query take the time to completeWhy? That answer can be totally meaningless as the VERY SAME query on the VERY SAME data on the VERY SAME database in the VERY SAME Oracle session can and will show DIFFERENT execution times.
    So why do you want to know a specific query's execution time? What do you expect that to tell you?
    If you mean that you want to know how long an existing query being executed is still going to take - that's usually quite difficult to determine. Oracle does provide a view on so-called long operations. However, only certain factors of a query's execution will trigger that this query is a long operation - and only for those specific queries will there be long operation stats that provide an estimated completion time.
    If your slow and long running query does not show in long operation, then Oracle does not consider it a long operation - it fails to meet the specific criteria and factors required as a long operation. This is not a bug or an error. Simply that your query does not meet the basic requirements to be viewed as a long operation.
    Oracle however provides the developer with the means to create long operations (using PL/SQL). You need to know and do the following:
    a) need to know how many units of work to do (e.g. how many fetches/loop iterations/rows your code will process)
    b) need to know how many units of work thus far done
    c) use the DBMS_APPLICATION_INFO package to create a long operation and continually update the operation with the number of work units thus far done
    It is pretty easy to implement this in PL/SQL processing code (assuming requirements a and b can be met) - and provide long operation stats and estimated completion time for the DBA/operators/users of the database, waiting on your process to complete.

  • Regarding Performance pbl in SQL query.

    hey guys,
    Could somebody help me over this posting.
    I have created TAX report. which retrieved records from BSIS and BSAS table to calculate consumption tax.
    From SE30 and SQL trace i found that loop at BSIS and BSAS takes round trip
    for each Input Period value.
    could you please give me best coding alternate to my below coding part.
    *Main code--
    loop at itab_t001.(itab_t001-bukrs is company code list,for each company code)
    WK_MONAT = '01'.
      DO.
        IF WK_MONAT = P_MONAT. "p_monat is parameter input for period
          FLG_CUR_PERIOD = CNS_TRUE.
        ENDIF.
        PERFORM FRM_READ_DATA.         " RETRIEVE DATA FROM BSIS AND BSAS
        WK_MONAT = WK_MONAT + 1.
        IF WK_MONAT > P_MONAT.
          EXIT.
        ENDIF.
      ENDDO.
    END of Main code----
          Read FI documents from table BSIS/BSAS
    FORM  FRM_READ_DATA.
    Select data from table BSIS
      SELECT BELNR BUZEI BLART MWSKZ HKONT SHKZG DMBTR
        INTO CORRESPONDING FIELDS OF TABLE ITAB_BSIS
        FROM BSIS WHERE BUKRS = itab_t001-BUKRS AND  GJAHR = P_GJAHR AND  MONAT = WK_MONAT AND
        ( MWSKZ <> ' ' AND MWSKZ <> '**' ).
    Select data from table BSAS
      SELECT BELNR BUZEI BLART MWSKZ HKONT SHKZG DMBTR
        APPENDING CORRESPONDING FIELDS OF TABLE ITAB_BSIS
        FROM BSAS  WHERE BUKRS = itab_t001-BUKRS AND  GJAHR = P_GJAHR AND MONAT = WK_MONAT AND
             ( MWSKZ <> ' ' AND MWSKZ <> '**' ).
    Select data from table BSIS that includes non-consumption tax accounts
      SELECT BELNR
       INTO TABLE ITAB_NO_TAX_DOC
        FROM BSIS WHERE BUKRS = itab_t001-BUKRS AND GJAHR = P_GJAHR AND
             MONAT = WK_MONAT AND HKONT IN S_NT_ACC.
    Select data from table BSAS that includes non-consumption tax accounts
      SELECT BELNR  APPENDING TABLE ITAB_NO_TAX_DOC
        FROM BSAS WHERE BUKRS = itab_t001-BUKRS AND  GJAHR = P_GJAHR AND
             MONAT = WK_MONAT AND HKONT IN S_NT_ACC.
    ...Proceeded with ABOVE fetched internal tables.
    ENDFORM.
    Note:
    I found that BSIS and BSAS table have 10000s of records it takes 3hrs time to run this report(if i give 12months period). (for each month(ie 01month report takes 15minutes to execute.)
    so i think instead calling BSiS twice, calling BSIS ones is wise.
    could you pls help me how can i modify the code and should i have
    to created any index here(if so mention how and which field)
    pls feel free to ask if any addition code is required.
    could you pls confirm.
    ambichan.
    Message was edited by: ambi chan

    Hi Ambichan,
    Few things that can be done to the below code.
    1. Dont put select queries inside the "loop at itab_t001....endloop" statement.
    Instead use <b>FOR ALL ENTRIES</b> IN itab_t001 for the queries. Then you can use READ stmt to get the values.
    NOTE: Make sure that you select all <b>key fields</b> from the databse when using FOR all Entries.
    <i>This will avoid hitting of the DB table multiple time.</i>
    2. Avoid "INTO CORRESPONDING FIELDS OF TABLE" in select queries. Modify the internal table to suit the strucutre of the databse table.
    <i>This will make the query faster</i>
    3. Make sure that the fields in the select query are in the same order as that in DB table.
    <i>This will also make the query faster</i>
    Try these changes and performance should increase.
    Regards,
    Saji.

  • How to measure the performance of a SQL query?

    Hello,
    I want to measure the performance of a group of SQL queries to compare them, but i don't know how to do it.
    Is there any application to do it?
    Thanks.

    You can use STATSPACK (in 10g its called as AWR - Automatic Workload Repository)
    Statspack -> A set of SQL, PL/SQL, and SQL*Plus scripts that allow the collection, automation, storage, and viewing of performance data. This feature has been replaced by the Automatic Workload Repository.
    Automatic Workload Repository - Collects, processes, and maintains performance statistics for problem detection and self-tuning purposes
    Oracle Database Performance Tuning Guide - Automatic Workload Repository
    http://download-east.oracle.com/docs/cd/B19306_01/server.102/b14211/autostat.htm#PFGRF02601
    or
    you can use EXPLAIN PLAN
    EXPLAIN PLAN -> A SQL statement that enables examination of the execution plan chosen by the optimizer for DML statements. EXPLAIN PLAN causes the optimizer to choose an execution plan and then to put data describing the plan into a database table.
    Oracle Database Performance Tuning Guide - Using EXPLAIN PLAN
    http://download-east.oracle.com/docs/cd/B19306_01/server.102/b14211/ex_plan.htm#PFGRF009
    Oracle Database SQL Reference - EXPLAIN PLAN
    http://download-east.oracle.com/docs/cd/B19306_01/server.102/b14200/statements_9010.htm#sthref8881

  • Performance of an sql query with 16 million rows

    Hello,
    I have about 16 million records in a project table, lets call it the ENCOUNTER table.
    I have an ENCOUNTER_NUMBER to be searched for which is VARCHAR2(30) and has a unique constraint on it. The results are to be ordered by ENCOUNTER_NUMBER.
    The search string that the user enters is always considered to be a contains search and so a % is prefixed and suffixed to the search string.
    I have created a functional index on UPPER(ENCOUNTER_NUMBER)
    More the number of matches to the search query, the less the time taken. If there are 100 matches or more, the query returns within a second.
    If there is only one match it takes atleast 10 seconds.
    I am not sure at this stage why there is huge difference. Our client would be happy with 5 seconds.
    Can anyone help me with this please, if at all it is possible to get the search quicker
    Any help would be greatly appreciated.
    Many thanks
    Marilyn

    You are seeing differing response times to fetch the first n rows - not all the rows. So if your search is 'less selective' then you need to search fewer rows to find the first n rows meeting your criteria.
    For example, if you have a search that 1% of your rows will match, then you probably need to scan about 10,000 rows to find the first 100 matching rows. If you specify a more selective predicate, matching 0.1% of the rows, then you will need to scan 100k rows to find your first 100 matches.
    A generic 'contains' query, such as you describe, is difficult to implement with good performance. And it is often the result of someone not giving sufficient thought to the requirements: 'Well, it's too hard to figure out how people really need to use it, so I'll ask for this and I'll be covered regardless of what people really need.'
    Can you negotiate/narrow down the requirements, such as:
    - allow leading wildcards or trailing wildcards but not both
    - allow/require other search criteria to be specified (not more double-wildcarded criteria, though)
    - allow searching specific positions of the column (for example, if positions 4-6 have some specific meaning, then allow a search on that substring)

  • Performance Tunning of Sql query

    Hi,
    I have table A with size 120 Million and two more tables are of size 2 Million on Table B and less than 1 Million size on table C.
    I had created Partition and Parallel degree 4 on the table A. Created table B with Parallel degree 2 and Created table C with NOPARALLEL.
    My query is using above tables with joins and inserting into table D using HINT /*+ APPEND NOLOGGING*/
    I had executed the explain on the above criteria the cost is showing 20767.
    Later created Tables A,B and C with NOPARALLEL. Applied HINT on Table D /*+ APPEND NOLOGGING*/
    and als applied HINT /*+ PARALLEL(A, 4) PARALLEL(B C, 2) on select query which uses to insert into Table D.
    My question which is best practice on PARALLEL degree creation at table level or query level:
    a) Creating table with Paralle (degree 4)
    b) Applying HINT /*+ PARALLEL (TABLE A , 4) */ at query level
    Regards,
    Prakash

    957901 wrote:
    Hi,
    I have table A with size 120 Million and two more tables are of size 2 Million on Table B and less than 1 Million size on table C.
    I had created Partition and Parallel degree 4 on the table A. Created table B with Parallel degree 2 and Created table C with NOPARALLEL.
    My query is using above tables with joins and inserting into table D using HINT /*+ APPEND NOLOGGING*/
    I had executed the explain on the above criteria the cost is showing 20767. Cost is meaningless out of context.
    Later created Tables A,B and C with NOPARALLEL. Applied HINT on Table D /*+ APPEND NOLOGGING*/
    and als applied HINT /*+ PARALLEL(A, 4) PARALLEL(B C, 2) on select query which uses to insert into Table D.
    My question which is best practice on PARALLEL degree creation at table level or query level:
    a) Creating table with Paralle (degree 4)
    b) Applying HINT /*+ PARALLEL (TABLE A , 4) */ at query level Whichever works best for what you're needing to achieve.
    Which is another way of saying you haven't provided enough information for anyone here to make any informed decision or suggestion for you.
    {message:id=9360002}
    For performance issues/tuning requests, read the two threads linked to by this FAQ: {message:id=9360003}

  • How to improve performance of this SQL query?

    Hi,
    I have a query that tries to build a string (RPATH) for use as a url parameter. The query is:
    SELECT DISTINCT USERNAME, PASSWORD, ROLE, RIGHTS,
    DECODE(GEO, ROLE, 'g='||NVL(GEO,'none'), NULL)||
         DECODE(AREA, ROLE, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none'), NULL)||
         DECODE(REGION, ROLE, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none')||'&r='||NVL(REGION,'none'), NULL)||
         DECODE(DISTRICT, ROLE, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none')||'&r='||NVL(REGION,'none')||'&d='||NVL(DISTRICT,'none'), NULL)||
         DECODE(OFFICE, ROLE, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none')||'&r='||NVL(REGION,'none')||'&d='||NVL(DISTRICT,'none')||'&o='||NVL(OFFICE,'none')
    , NULL) RPATH
    FROM (SELECT U.*, L.*
         FROM (SELECT * FROM T_USERS WHERE USERNAME='xxx' AND PASSWORD='yyy') U, T_LOC_SUB L
         WHERE U.ROLE IN ('WW', L.GEO, L.AREA, L.REGION, L.DISTRICT, L.OFFICE))
    GROUP BY USERNAME, PASSWORD, ROLE, RIGHTS, GEO, AREA, REGION, DISTRICT, OFFICE;
    T_USERS is defined as
    CREATE TABLE T_USERS (
    username          VARCHAR2(10)     CONSTRAINT T_USERS_username_pk PRIMARY KEY,
    password          VARCHAR2(10),
    role                    CONSTRAINT T_USERS_role_FK REFERENCES T_LOC_MAIN(loc),
    rights          VARCHAR2(3)
    T_LOC_SUB is defined as
    CREATE TABLE T_LOC_SUB (
    geo          CONSTRAINT T_LOC_SUB_geo_FK REFERENCES T_LOC_MAIN(loc),
    area          CONSTRAINT T_LOC_SUB_area_FK REFERENCES T_LOC_MAIN(loc),
    region          CONSTRAINT T_LOC_SUB_region_FK REFERENCES T_LOC_MAIN(loc),
    district     CONSTRAINT T_LOC_SUB_district_FK REFERENCES T_LOC_MAIN(loc),
    office          CONSTRAINT T_LOC_SUB_office_FK REFERENCES T_LOC_MAIN(loc)
    T_LOC_MAIN is defined as
    CREATE TABLE T_LOC_MAIN (
    loc          VARCHAR2(4)     CONSTRAINT T_LOC_MAIN_loc_PK PRIMARY KEY,
    label          VARCHAR2(60),
    rank          NUMBER
    REGION and DISTRICT columns in T_LOC_SUB may be NULL at times. How can I rewrite the SQL to make it run faster or more efficiently?
    Please help.. Thank you..

    Hi,
    I just realised I can simplify the query to:
    SELECT DISTINCT USERNAME, PASSWORD, ROLE, RIGHTS,
    DECODE(ROLE,
         GEO, 'g='||NVL(GEO,'none'),
         AREA, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none'),
         REGION, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none')||'&r='||NVL(REGION,'none'),
         DISTRICT, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none')||'&r='||NVL(REGION,'none')||'&d='||NVL(DISTRICT,'none'),
         OFFICE, 'g='||NVL(GEO,'none')||'&a='||NVL(AREA,'none')||'&r='||NVL(REGION,'none')||'&d='||NVL(DISTRICT,'none')||'&o='||NVL(OFFICE,'none'),
         NULL) RPATH
    FROM (SELECT U.*, L.*
    FROM (SELECT * FROM T_USERS WHERE USERNAME='xxx' AND PASSWORD='yyy') U, T_LOC_SUB L
    WHERE U.ROLE IN ('WW', L.GEO, L.AREA, L.REGION, L.DISTRICT, L.OFFICE))
    GROUP BY USERNAME, PASSWORD, ROLE, RIGHTS, GEO, AREA, REGION, DISTRICT, OFFICE;
    Anyone can offer a better and more efficient improvement?
    Thanx!

  • Performance Issue with sql query

    Hi,
    My db is 10.2.0.5 with RAC on ASM, Cluster ware version 10.2.0.5.
    With bsoa table as
    SQL> desc bsoa;
    Name                                      Null?    Type
    ID                                        NOT NULL NUMBER
    LOGIN_TIME                                         DATE
    LOGOUT_TIME                                        DATE
    SUCCESSFUL_IND                                     VARCHAR2(1)
    WORK_STATION_NAME                                  VARCHAR2(80)
    OS_USER                                            VARCHAR2(30)
    USER_NAME                                 NOT NULL VARCHAR2(30)
    FORM_ID                                            NUMBER
    AUDIT_TRAIL_NO                                     NUMBER
    CREATED_BY                                         VARCHAR2(30)
    CREATION_DATE                                      DATE
    LAST_UPDATED_BY                                    VARCHAR2(30)
    LAST_UPDATE_DATE                                   DATE
    SITE_NO                                            NUMBER
    SESSION_ID                                         NUMBER(8)
    The query
    UPDATE BSOA SET LOGOUT_TIME =SYSDATE WHERE SYS_CONTEXT('USERENV', 'SESSIONID') = SESSION_ID
    Is taking a lot of time to execute and in AWR reports also it is on top in
    1. SQL Order by elapsed time
    2. SQL order by reads
    3. SQL order by gets
    So i am trying a way to solve the performance issue as the application is slow specially during login and logout time.
    I understand that the function in the where condition cause to do FTS, but i can not think what other parts to look at.
    Also:
    SQL> SELECT COUNT(1) FROM BSOA;
      COUNT(1)
       7800373
    The explain plan for  "UPDATE BSOA SET LOGOUT_TIME =SYSDATE WHERE SYS_CONTEXT('USERENV', 'SESSIONID') = SESSION_ID" is
    {code}
    PLAN_TABLE_OUTPUT
    Plan hash value: 1184960901
    | Id  | Operation          | Name               | Rows  | Bytes | Cost (%CPU)| Time     |
    |   0 | UPDATE STATEMENT   |                    |     1 |    26 | 18748   (3)| 00:03:45 |
    |   1 |  UPDATE            | BSOA |       |       |            |          |
    |*  2 |   TABLE ACCESS FULL| BSOA |     1 |    26 | 18748   (3)| 00:03:45 |
    Predicate Information (identified by operation id):
       2 - filter("SESSION_ID"=TO_NUMBER(SYS_CONTEXT('USERENV','SESSIONID')))
    {code}

    Hi,
    There are also triggers before update and AUDITS on this table.
    CREATE OR REPLACE TRIGGER B2.TRIGGER1
    BEFORE UPDATE
    ON B2.BSOA  REFERENCING OLD AS OLD NEW AS NEW
    FOR EACH ROW
    :NEW.LAST_UPDATED_BY   := USER    ;
    :NEW.LAST_UPDATE_DATE  := SYSDATE ;
    END;
    CREATE OR REPLACE TRIGGER B2.TRIGGER2
    BEFORE INSERT
    ON B2.BSOA  REFERENCING OLD AS OLD NEW AS NEW
    FOR EACH ROW
    :NEW.CREATED_BY        := USER ;
    :NEW.CREATION_DATE     := SYSDATE ;
    :NEW.LAST_UPDATED_BY   := USER    ;
    :NEW.LAST_UPDATE_DATE  := SYSDATE ;
    END;
    And also there is an audit on this table
    AUDIT UPDATE ON B2.BSOA BY ACCESS WHENEVER SUCCESSFUL;
    AUDIT UPDATE ON B2.BSOA BY ACCESS WHENEVER NOT SUCCESSFUL;
    And the sessionid column in BSOA has height balanced histogram.
    When i create an index i get the following error. As i am on 10g I can't use DDL_LOCK_TIMEOUT . I may have to wait for next down time.
    SQL> CREATE INDEX B2.BSOA_SESSID_I ON B2.BSOA(SESSION_ID) TABLESPACE B2 COMPUTE STATISTICS;
    CREATE INDEX B2.BSOA_SESSID_I ON B2.BSOA(SESSION_ID) TABLESPACE B2 COMPUTE STATISTICS
    ERROR at line 1:
    ORA-00054: resource busy and acquire with NOWAIT specified
    Thanks

  • Performance tuning in SQL query using join of views

    Hi,
    Am trying to tune a query of the format
    select ........ from view1,view2
    where view1.keyfield = view2.keyfield
    The base tables of the views view1 and view2 have indexes specified on 'keyfield'.
    However, when I do an explain plan of the query, Full table scan of those base tables are performed.
    Even tried using Hints in this query to force Oracle to use those indexes, but still not successful.
    Any pointers on how to tune this kind of query would be highly welcome.
    Regards,
    Baish

    If your query is really of the same form as you posted, then full table scans then a hash or merge join may be the most efficient plan.
    You are asking for all of the rows from view1 and the matching rows from view2, because of this, Oracle dedcided that the cost of full table scans using multi-block reads is cheaper than using single block reads to get the rowids from the index, then single block reads to get the rows from the table.
    FULL SCAN <> BAD
    John

  • Performance tuning of sql query with multiple joins

    My query takes at least half an hour to execute and the number of records returned are approx 1 lakh records.
    Structure of tables are:
    tblSession : ID,option1,option2,option3,option4,option5,option6,option7,option8,option9.
    tblOption : ID, labelID
    tblLabelDetail  : ID, LABELID, text
    optionID 1 to optionID9 are Foreign keys to table tblOption.ID
    My query is as below : 
    select 
    session.ID 
    ,session.tstamp
    ,session.score
    ,session.hid1
    ,session.hID2
    ,session.hID3
    ,session.collectionID
    ,session.consumerID
    ,session.langID
    ,cons_cust.text1 as    customCons_text1, 
    cons_cust.text2 as customCons_text2, 
    cons_cust.text3 as customCons_text3,
    cons_cust.text4 as customCons_text4,
    cons_cust.text5 as customCons_text5,
    cons_cust.text6 as customCons_text6,
    cons_cust.text7 as customCons_text7,
    cons_cust.text8 as customCons_text8,
    cons_cust.text9 as customCons_text9,
    ld_cons1.text as customCons_option1GUID, 
    ld_cons2.text as customCons_option2GUID, 
    ld_cons3.text as customCons_option3GUID, 
    ld_cons4.text as customCons_option4GUID ,
    ld_cons5.text as customCons_option5GUID, 
    ld_cons6.text as customCons_option6GUID, 
    ld_cons7.text as customCons_option7GUID, 
    ld_cons8.text as customCons_option8GUID, 
    ld_cons9.text as customCons_option9GUID,
    --session
    session_cust.text1 as  session_cust_text1, 
    session_cust.text2 as session_cust_text2, 
    session_cust.text3 as session_cust_text3,
    session_cust.text4 as session_cust_text4,
    session_cust.text5 as session_cust_text5,
    session_cust.text6 as session_cust_text6,
    session_cust.text7 as session_cust_text7,
    session_cust.text8 as session_cust_text8,
    session_cust.text9 as session_cust_text9,
    ld_sess1.text as session_cust_option1GUID, 
    ld_sess2.text as session_cust_option2GUID, 
    ld_sess3.text as session_cust_option3GUID, 
    ld_sess4.text as session_cust_option4GUID, 
    ld_sess5.text as session_cust_option5GUID, 
    ld_sess6.text as session_cust_option6GUID, 
    ld_sess7.text as session_cust_option7GUID, 
    ld_sess8.text as session_cust_option8GUID, 
    ld_sess9.text as session_cust_option9GUID, 
    session_cust.tStamp1,
    session_cust.tStamp2
    from mvSession session with (noexpand)
    inner join tblCollection c on c.ID=session.collectionID AND c.templateID = 405
    left join tblConsumer cons on cons.ID=session.consumerID and cons.sessionYM between 601 and 1412 and cons.sID=105
    left join vCustomConsumer cons_cust on cons_cust.sessionYM between 601 and 1412 and cons_cust.sID=105 and cons_cust.ID=cons.ID
    left join tbloption o_cons1 on o_cons1.id = cons_cust.option1 and  o_cons1.sid = 105
    left join tblLabelDetail ld_cons1 on ld_cons1.labelID = o_cons1.labelID and ld_cons1.langId = 1 and ld_cons1.eid = 107 
    left join tbloption o_cons2 on o_cons2.id = cons_cust.option2 and  o_cons2.sid = 105
    left join tblLabelDetail ld_cons2 on ld_cons2.labelID = o_cons2.labelID and ld_cons2.langId = 1 and ld_cons2.eid = 107 
    left join tbloption o_cons3 on o_cons3.id = cons_cust.option3 and  o_cons3.sid = 105
    left join tblLabelDetail ld_cons3 on ld_cons3.labelID = o_cons1.labelID and ld_cons3.langId = 1 and ld_cons3.eid = 107 
    left join tbloption o_cons4 on o_cons4.id = cons_cust.option4 and  o_cons4.sid = 105
    left join tblLabelDetail ld_cons4 on ld_cons4.labelID = o_cons4.labelID and ld_cons4.langId = 1 and ld_cons4.eid = 107 
    left join tbloption o_cons5 on o_cons5.id = cons_cust.option5 and  o_cons5.sid = 105
    left join tblLabelDetail ld_cons5 on ld_cons5.labelID = o_cons5.labelID and ld_cons5.langId = 1 and ld_cons5.eid = 107 
    left join tbloption o_cons6 on o_cons6.id = cons_cust.option6 and  o_cons6.sid = 105
    left join tblLabelDetail ld_cons6 on ld_cons6.labelID = o_cons6.labelID and ld_cons6.langId = 1 and ld_cons6.eid = 107 
    left join tbloption o_cons7 on o_cons7.id = cons_cust.option7 and  o_cons7.sid = 105
    left join tblLabelDetail ld_cons7 on ld_cons7.labelID = o_cons7.labelID and ld_cons7.langId = 1 and ld_cons7.eid = 107 
    left join tbloption o_cons8 on o_cons8.id = cons_cust.option8 and  o_cons8.sid = 105
    left join tblLabelDetail ld_cons8 on ld_cons8.labelID = o_cons8.labelID and ld_cons8.langId = 1 and ld_cons8.eid = 107 
    left join tbloption o_cons9 on o_cons9.id = cons_cust.option9 and  o_cons9.sid = 105
    left join tblLabelDetail ld_cons9 on ld_cons9.labelID = o_cons9.labelID and ld_cons9.langId = 1 and ld_cons9.eid = 107 
    left join vCustomSession session_cust on session_cust.sessionYM between 601 and 1412 and session_cust.sID=105 and session_cust.ID=session.ID
    left join tbloption o_sess1 on o_sess1.id = session_cust.option1 and  o_sess1.sid = 105
    left join tblLabelDetail ld_sess1 on ld_sess1.labelID = o_sess1.labelID and ld_sess1.langId = 1 and ld_sess1.eid = 107 
    left join tbloption o_sess2 on o_sess2.id = session_cust.option2 and  o_sess2.sid = 105
    left join tblLabelDetail ld_sess2 on ld_sess2.labelID = o_sess2.labelID and ld_sess2.langId = 1 and ld_sess2.eid = 107 
    left join tbloption o_sess3 on o_sess2.id = session_cust.option3 and  o_sess3.sid = 105
    left join tblLabelDetail ld_sess3 on ld_sess3.labelID = o_sess2.labelID and ld_sess3.langId = 1 and ld_sess3.eid = 107 
    left join tbloption o_sess4 on o_sess4.id = session_cust.option4 and  o_sess4.sid = 105
    left join tblLabelDetail ld_sess4 on ld_sess4.labelID = o_sess4.labelID and ld_sess4.langId = 1 and ld_sess4.eid = 107 
    left join tbloption o_sess5 on o_sess5.id = session_cust.option5 and  o_sess5.sid = 105
    left join tblLabelDetail ld_sess5 on ld_sess5.labelID = o_sess5.labelID and ld_sess5.langId = 1 and ld_sess5.eid = 107 
    left join tbloption o_sess6 on o_sess6.id = session_cust.option6 and  o_sess6.sid = 105
    left join tblLabelDetail ld_sess6 on ld_sess6.labelID = o_sess6.labelID and ld_sess6.langId = 1 and ld_sess6.eid = 107 
    left join tbloption o_sess7 on o_sess7.id = session_cust.option7 and  o_sess7.sid = 105
    left join tblLabelDetail ld_sess7 on ld_sess7.labelID = o_sess7.labelID and ld_sess7.langId = 1 and ld_sess7.eid = 107 
    left join tbloption o_sess8 on o_sess8.id = session_cust.option8 and  o_sess8.sid = 105
    left join tblLabelDetail ld_sess8 on ld_sess8.labelID = o_sess8.labelID and ld_sess8.langId = 1 and ld_sess8.eid = 107 
    left join tbloption o_sess9 on o_sess9.id = session_cust.option9 and  o_sess9.sid = 105
    left join tblLabelDetail ld_sess9 on ld_sess9.labelID = o_sess9.labelID and ld_sess9.langId = 1 and ld_sess9.eid = 107 
    where session.sID=105  and session.tStamp >= 'Sep  1 2014 12:00AM' and session.tStamp < 'Dec 12 2014 12:00AM'   
    order by session.tStamp, session.ID
    Is there a way , where i can simplify the joins with tbloption and tblLabelDetail and get my o/p in optimal time.
    Regards 

    I have headed towards another approach ie. using unpivot and then pivot.
    First i am converting option1-option9 into column , then doing  PIVOT to get back the same record . But issue is that when i am doing pivoting i am getting NULL values.
    My query is :
    select * into #t1  from
    select ID
    , option1
    , option2
    , option3
    , option4
    , option5
    , option6
    , option7
    , option8
    , option9
    from vCustomConsumer
    where sid=105
    and sessionYM = 1412 
    ) SourceTable
    UNPIVOT
       optionID FOR Col IN
        (option1 
    ,option2
    ,option3 
    ,option4 
    ,option5 
    ,option6 
    ,option7 
    ,option8 
    ,option9 )
    ) AS unpvt
    select t.ID,t.optionID,t.col,cast(ld.text as varchar(max)) as text into #t2
    from #t1 t
    left outer join tbloption o on o.ID =  t.optionID
    left outer join tblLabelDetail ld on ld.labelID = o.labelID and ld.langID=1 
    order by ID,col
    select ID,option1 
    ,option2
    ,[option3]
    ,option4 
    ,option5 
    ,option6 
    ,option7 
    ,option8 
    ,option9
    from
    select ID,optionID,col,text
    from #t2
    )up
    pivot 
    min(text)for col in 
    (option1 
    ,option2
    ,[option3]
    ,option4 
    ,option5 
    ,option6 
    ,option7 
    ,option8 
    ,option9
    )as pvt
    In my last query where i am using pivot, i am getting NULL values. When i check the data in temp table #t2 , it exists perfectly . But when pivoting i dont understand why it is returning most of the NULL values. I am getting data for only one column in single
    row.
    Below are some rows from result set finally obtained after pivoting :
    ID
    option1
    option2
    option3
    option4
    option5
    option6
    option7
    option8
    option9
    62949026
    NULL
    0 to 200 seconds
    NULL
    NULL
    NULL
    NULL
    NULL
    NULL
    NULL
    62966000
    NULL
    NULL
    4
    NULL
    NULL
    NULL
    NULL
    NULL
    NULL
    62966032
    NULL
    NULL
    4
    NULL
    NULL
    NULL
    NULL
    NULL
    NULL
    63090372
    NULL
    NULL
    NULL
    NULL
    EN
    NULL
    NULL
    NULL
    NULL
    63090375
    NULL
    NULL
    NULL
    NULL
    EN
    NULL
    NULL
    NULL
    NULL
    Thanks,

Maybe you are looking for

  • Can I use more then one apple ID on my computer/osx app store

    Hi, I use my MBP/iMac/ipad/iphone for both business and home.  Right now everything is setup to my home apple id (and I have made quite a few purchases with that).  With the release of the osx app store and FCPX,  my company wants to buy these for my

  • How can i add multiple fields when multiple fields match?

    Ok, I am really new to JavaScript and need as much help as I can get. Ok, here's what I have: I have 10 drop boxes labeled HAZ1-HAZ10. In each drop box the user can choose 1.1 - 1.6. I also have 10 numeric fields labeled NEW1T - NEW10T. These numeric

  • Wipe iPhone

    How do I wipe my phone before passing to another soul

  • Components getting cut off on resizing

    I have a custom panel that sits within a JViewport that sits within a JScrollPane. As I increase the size of the panel, the scrollbars allow me to scroll up and down to see the new area. However, anything I try to put in the area below the original 2

  • OLTP to Star

    I have three physical tables: GL_BALANCES (columns: corp_acct_id, amount) CORP_ACCOUNTS (columns: corp_acct_id) INTERFACE_DETAIL (columns: corp_acct_id, source_system_acct_id, source_system_sub_acct_id, amount) There is a one to many relationship bet