Select statement Table Join sequence (Performance)

Hi,
I have 3 tables Header (Hdr), Item (Itm) and Schedule Line (Sdl) table. The number of entries is max in Sdl table (Eg: Hdr = 1000; Item = 20000 and Sdl = 200000 records). I need to join these 3 tables in my SELECT statement to fetch the required records.
Case A:
Header AS HDR
JOIN Item AS ITM
ON HDRORDID = ITMORDID
AND HDRVRSIOID = ITMVRSIOID
JOIN ScheduleLine AS SDL
ON ITMITMID = SDLITMID
Case B:
ScheduleLine AS SDL
JOIN Item AS ITM
ON ITMITMID = SDLITMID
JOIN Header AS HDR
ON HDRORDID = ITMORDID
AND HDRVRSIOID = ITMVRSIOID
Questions:
1) The performance of the Select query is really bad with Case A and very good (really fast) with Case B. Will the number of records in the tables used determine the sequence of join in the Select statement ?
2) If yes, is this independent of database used ? That is, will the behaviour be the same in all databases (Eg: Oracle, DB2, SQL Server, etc) ?
Appreciate your help.
Thanks,
Karthik

I believe that the performance discrepancy between the two versions is caused by you not specifying the joins correctly (i.e. not qualifying the header and item at schedule level as I mentioned previously).
CAse A.
Header AS HDR
JOIN Item AS ITM
ON HDRORDID = ITMORDID
AND HDRVRSIOID = ITMVRSIOID
JOIN ScheduleLine AS SDL
ON ITMITMID = SDLITMID
reads all 1000 hDRs, joins all 20000 items correctly to these BUT then joins schedule lines only by item id so there will be a very large number of matches. i.e. any item with itemid 10 will match to EVERY SDL with itemid 10. as there are 20000 items and 200000 schedule lines this will be massive!
Case B:
ScheduleLine AS SDL
JOIN Item AS ITM
ON ITMITMID = SDLITMID
JOIN Header AS HDR
ON HDRORDID = ITMORDID
AND HDRVRSIOID = ITMVRSIOID
This will not have the same amount of DB traffic as the SDL is now the driving force. It will still match more than you intend but will only be reading the items redundantly trather than the more voluminous schedules.
Basically, your sql is wrong and should qualify the matches right down to schedule line level as per my earlier post. When you have changed the sql, try the two alternatives again and see if there is any performance differences. I suspect that there will not be, but it will be interesting to find out for sure.

Similar Messages

Assign a 'Linked Selection' state to a Sequence?

I have two Sequences in a Project. In one of the Sequences I am chopping out bits of both Video and Audio but in the other I want to only chop out bits of Video and leave the Audio alone. Currently I'm having to click on 'Linked Selection' as I switch between the Sequences. Is there a way of assigning a particular state of the 'Linked Selection' with a specific Sequence?
Thanks in advance.

The Linked Sekection button is for the window not the sequence. You could select the clips in the sequence you don't want linked selection and use Cmd-L to unlink video and audio.

Joins And For all Enteries in Select Statement

Could you please tell me when there is a high amount of data which is being handled in the table, does the use of INNER JOINS and FOR ALL ENTERIES in SELECT Statement decreases the system performance? ?
Can you also let me know where can i get some tips regarding do's and dont's for ABAP Programming, I want to increase my system performance.
Currently the programs which are being used are taking a lot of time for execution...
Its very URGENT!

Hai Jyotsna
Go through the following Tips for improving Performence
For all entries
The for all entries creates a where clause, where all the entries in the driver table are combined with OR. If the number of entries in the driver table is larger than rsdb/max_blocking_factor, several similar SQL statements are executed to limit the length of the WHERE clause.
The plus
Large amount of data
Mixing processing and reading of data
Fast internal reprocessing of data
Fast
The Minus
Difficult to program/understand
Memory could be critical (use FREE or PACKAGE size)
Some steps that might make FOR ALL ENTRIES more efficient:
Removing duplicates from the driver table
Sorting the driver table
If possible, convert the data in the driver table to ranges so a BETWEEN statement is used instead of and OR statement:
               FOR ALL ENTRIES IN i_tab
                  WHERE mykey >= i_tab-low and
             mykey <= i_tab-high.
Nested selects
The plus:
Small amount of data
Mixing processing and reading of data
Easy to code - and understand
The minus:
Large amount of data
when mixed processing isnt needed
Performance killer no. 1
Select using JOINS
The plus
Very large amount of data
Similar to Nested selects - when the accesses are planned by the programmer
In some cases the fastest
Not so memory critical
The minus
Very difficult to program/understand
Mixing processing and reading of data not possible
Use the selection criteria
SELECT * FROM SBOOK.
CHECK: SBOOK-CARRID = 'LH' AND
                  SBOOK-CONNID = '0400'.
ENDSELECT.
SELECT * FROM SBOOK
WHERE CARRID = 'LH' AND
        CONNID = '0400'.
ENDSELECT.
Use the aggregated functions
C4A = '000'.
SELECT * FROM T100
WHERE SPRSL = 'D' AND
        ARBGB = '00'.
CHECK: T100-MSGNR > C4A.
C4A = T100-MSGNR.
ENDSELECT.
SELECT MAX( MSGNR ) FROM T100 INTO C4A
WHERE SPRSL = 'D' AND
       ARBGB = '00'.
Select with view
SELECT * FROM DD01L
WHERE DOMNAME LIKE 'CHAR%'
        AND AS4LOCAL = 'A'.
SELECT SINGLE * FROM DD01T
    WHERE   DOMNAME    = DD01L-DOMNAME
        AND AS4LOCAL   = 'A'
        AND AS4VERS    = DD01L-AS4VERS
        AND DDLANGUAGE = SY-LANGU.
ENDSELECT.
SELECT * FROM DD01V
WHERE DOMNAME LIKE 'CHAR%'
       AND DDLANGUAGE = SY-LANGU.
ENDSELECT.
Select with index support
SELECT * FROM T100
WHERE     ARBGB = '00'
       AND MSGNR = '999'.
ENDSELECT.
SELECT * FROM T002.
SELECT * FROM T100
    WHERE     SPRSL = T002-SPRAS
          AND ARBGB = '00'
          AND MSGNR = '999'.
ENDSELECT.
ENDSELECT.
Select Into table
REFRESH X006.
SELECT * FROM T006 INTO X006.
APPEND X006.
ENDSELECT
SELECT * FROM T006 INTO TABLE X006.
Select with selection list
SELECT * FROM DD01L
WHERE DOMNAME LIKE 'CHAR%'
        AND AS4LOCAL = 'A'.
ENDSELECT
SELECT DOMNAME FROM DD01L
INTO DD01L-DOMNAME
WHERE DOMNAME LIKE 'CHAR%'
       AND AS4LOCAL = 'A'.
ENDSELECT
Key access to multiple lines
LOOP AT TAB.
CHECK TAB-K = KVAL.
ENDLOOP.
LOOP AT TAB WHERE K = KVAL.
ENDLOOP.
Copying internal tables
REFRESH TAB_DEST.
LOOP AT TAB_SRC INTO TAB_DEST.
APPEND TAB_DEST.
ENDLOOP.
TAB_DEST[] = TAB_SRC[].
Modifying a set of lines
LOOP AT TAB.
IF TAB-FLAG IS INITIAL.
    TAB-FLAG = 'X'.
ENDIF.
MODIFY TAB.
ENDLOOP.
TAB-FLAG = 'X'.
MODIFY TAB TRANSPORTING FLAG
           WHERE FLAG IS INITIAL.
Deleting a sequence of lines
DO 101 TIMES.
DELETE TAB_DEST INDEX 450.
ENDDO.
DELETE TAB_DEST FROM 450 TO 550.
Linear search vs. binary
READ TABLE TAB WITH KEY K = 'X'.
READ TABLE TAB WITH KEY K = 'X' BINARY SEARCH.
Comparison of internal tables
DESCRIBE TABLE: TAB1 LINES L1,
                TAB2 LINES L2.
IF L1 <> L2.
TAB_DIFFERENT = 'X'.
ELSE.
TAB_DIFFERENT = SPACE.
LOOP
AT TAB1.
    READ TABLE TAB2 INDEX SY-TABIX.
    IF TAB1 <> TAB2.
      TAB_DIFFERENT = 'X'. EXIT.
    ENDIF.
ENDLOOP.
ENDIF.
IF TAB_DIFFERENT = SPACE.
ENDIF.
IF TAB1[] = TAB2[].
ENDIF.
Modify selected components
LOOP AT TAB.
TAB-DATE = SY-DATUM.
MODIFY TAB.
ENDLOOP.
WA-DATE = SY-DATUM.
LOOP AT TAB.
MODIFY TAB FROM WA TRANSPORTING DATE.
ENDLOOP.
Appending two internal tables
LOOP AT TAB_SRC.
APPEND TAB_SRC TO TAB_DEST.
ENDLOOP
APPEND LINES OF TAB_SRC TO TAB_DEST.
Deleting a set of lines
LOOP AT TAB_DEST WHERE K = KVAL.
DELETE TAB_DEST.
ENDLOOP
DELETE TAB_DEST WHERE K = KVAL.
Tools available in SAP to pin-point a performance problem
·                The runtime analysis (SE30)
·                SQL Trace (ST05)
·                Tips and Tricks tool
·                The performance database
Optimizing the load of the database
Using table buffering
Using buffered tables improves the performance considerably. Note that in some cases a statement can not be used with a buffered table, so when using these statements the buffer will be bypassed. These statements are:
Select DISTINCT
ORDER BY / GROUP BY / HAVING clause
Any WHERE clause that contains a sub query or IS NULL expression
JOIN s
A SELECT... FOR UPDATE
If you wan t to explicitly bypass the buffer, use the BYPASS BUFFER addition to the SELECT clause.
Use the ABAP SORT Clause Instead of ORDER BY
The ORDER BY clause is executed on the database server while the ABAP SORT statement is executed on the application server. The database server will usually be the bottleneck, so sometimes it is better to move the sort from the database server to the application server.
If you are not sorting by the primary key ( E.g. using the ORDER BY PRIMARY key statement) but are sorting by another key, it could be better to use the ABAP SORT statement to sort the data in an internal table. Note however that for very large result sets it might not be a feasible solution and you would want to let the database server sort it.
Avoid the SELECT DISTINCT Statement
As with the ORDER BY clause it could be better to avoid using SELECT DISTINCT, if some of the fields are not part of an index. Instead use ABAP SORT + DELETE ADJACENT DUPLICATES on an internal table, to delete duplicate rows.
Thanks & regards
Sreenivasulu P

ABAP Joins or Separate select statements?

Hi,
I am working on a simulation of an implementation project as part of my in-house training programme. I am a rookie to ABAP and am now doing a print program that requires master data selection from tables ANLZ, LFA1, MSEG and CSKT, and also related line items from ANLA.
As a former Java/.NET programmer I used to play with database joins. Now since ABAP joins are much simpler to use, I take the best advantage of them. For instance, in this particular program, I fetched related data from ANLZ, ANLA, LFA1 and CSKT into one internal table using a single select statement with ABAP joins.
Now I happen to hear (in bits and pieces) that one should limit the use of nested joins to a certain extend. Instead, for the above program, I was advised to use 4 internal tables and to select data separately using FOR ALL ENTRIES. But in this case, we need to use 4 select statements, which means 4 database operations in place of just one.
From ABAP documentation, I read that we can link upto 24 tables in a single select statement using ABAP joins.
Can anyone clarify more about this? Which is high on performance?
Using a single select statement using joins
or
Using multiple selects -- FOR ALL ENTRIES?
Thanks and regards,
Arun

Hi Arun,
you keep reading in these forums that FOR ALL ENTRIES is more performant, but this is simply not true.
In your example, joining four tables for one DB operation is better than storing a lot of redundant data in internal tables just to perform four separate DB operations using FOR ALL ENTRIES.
Make sure you join the tables correctly, i.e. link dependent tables giving their full primary key.
FOR ALL ENTRIES can be used where effective join statements are not possible, e.g. when cluster tables are involved (like BSEG).
Cheers
Thomas
Edit: check this out too: Inner Join or For All Entries

Problem with Select Statements

Hi All,
I have a performance problem for my report because of the following statements.
How can i modify the select statements for improving the performance of the report.
DATA : shkzg1h LIKE bsad-shkzg,
         shkzg1s LIKE bsad-shkzg,
         shkzg2h LIKE bsad-shkzg,
         shkzg2s LIKE bsad-shkzg,
         shkzg1hu LIKE bsad-shkzg,
         shkzg1su LIKE bsad-shkzg,
         shkzg2hu LIKE bsad-shkzg,
         shkzg2su LIKE bsad-shkzg,
         kopbal1s LIKE bsad-dmbtr,
         kopbal2s LIKE bsad-dmbtr,
         kopbal1h LIKE bsad-dmbtr,
         kopbal2h LIKE bsad-dmbtr,
         kopbal1su LIKE bsad-dmbtr,
         kopbal2su LIKE bsad-dmbtr,
         kopbal1hu LIKE bsad-dmbtr,
         kopbal2hu LIKE bsad-dmbtr.
*These statements are in LOOP.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg1s , kopbal1s)
      FROM bsid
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'S'
       AND umskz EQ ''
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg1su , kopbal1su)
      FROM bsid
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'S'
       AND umskz IN zspgl
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg1h , kopbal1h)
      FROM bsid
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'H'
       AND umskz EQ ''
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg1hu , kopbal1hu)
      FROM bsid
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'H'
       AND umskz IN zspgl
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg2s , kopbal2s)
      FROM bsad
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'S'
       AND umskz EQ ''
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg2su , kopbal2su)
      FROM bsad
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'S'
       AND umskz IN zspgl
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg2h , kopbal2h)
      FROM bsad
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'H'
       AND umskz EQ ''
     GROUP BY shkzg.
    ENDSELECT.
    SELECT shkzg SUM( dmbtr )
      INTO (shkzg2hu , kopbal2hu)
      FROM bsad
     WHERE bukrs = ibukrs
       AND kunnr = ktab-kunnr
       AND budat < idate-low
       AND shkzg = 'H'
       AND umskz IN zspgl
     GROUP BY shkzg.
    ENDSELECT.

>
Siegfried Boes wrote:
> Please stop writing answers if you understrand nothing about database SELECTS!
> All above recommendations are pure nonsense!
>
> As always with such questions, you must do an analysis before you ask! The coding itself is perfectly o.k., a SELECT with an aggregate and a GROUP BY can not be changed into a SELECT SINGLE or whatever.
>
> But your SELECTS mustr be supported by indexes!
>
> Please run SQL Trace, and tell us the results:
>
> I see 8 statements, what is the duration and the number of records coming back for each statement?
> Maybe only one statement is slow.
>
> See
> SQL trace:
> /people/siegfried.boes/blog/2007/09/05/the-sql-trace-st05-150-quick-and-easy
>
>
> Siegfried
Nice point there Siegfried. Instead of giving constructive suggestion, people here give a very bad suggestion on using SELECT SINGLE combined with SUM and GROUP BY.
I hope the person already look at your reply before he try using select single and wondering why he has error.
Anyway, the most important thing is how many loop expected for those select statements?
If you have like thousands of loop, you can expect a poor performance.
So, you should also look at how many times the select statement is called and not only performance for each select statement when you're doing SQL trace.
Regards,
Abraham

Performance problem(ANEA/ANEP table) in Select statement

Hi
I am using below select statement to fetch data.
Does the below where statement have performance issue?
can you Pls suggest.
1)In select of ANEP table, i am not using all the Key field in where condition. will it have performance problem?
2)does the order of where condition should be same as in table, if any one field order change also will have effect performance
SELECT bukrs
         anln1
         anln2
         afabe
         gjahr
         peraf
         lnran
         bzdat
         bwasl
         belnr
         buzei
         anbtr
         lnsan
    FROM anep
    INTO TABLE o_anep
    FOR ALL ENTRIES IN i_anla
   WHERE bukrs = i_anla-bukrs
     AND anln1 = i_anla-anln1
     AND anln2 = i_anla-anln2
     AND afabe IN s_afabe
     AND bzdat =< p_date
     AND bwasl IN s_bwasl.
SELECT bukrs
         anln1
         anln2
         gjahr
         lnran
         afabe
         aufwv
         nafal
         safal
         aafal
         erlbt
         aufwl
         nafav
         aafav
         invzv
         invzl
    FROM anea
    INTO TABLE o_anea
    FOR ALL ENTRIES IN o_anep
   WHERE bukrs = o_anep-bukrs
     AND anln1 = o_anep-anln1
     AND anln2 = o_anep-anln2
     AND gjahr = o_anep-gjahr
     AND lnran = o_anep-lnran
     AND afabe = o_anep-afabe.
Moderator message: Please Read before Posting in the Performance and Tuning Forum
Edited by: Thomas Zloch on Aug 9, 2011 9:37 AM

1. Yes. If you have only a few primary keys in youe WHERE condition that does affect the performance. But some times requirement itself may be in that way. We may not be knowing all the primary keys to given them in WHER conditon. If you know the values, then provide them without fail.
2. Yes. It's better to always follow the sequence in WHERE condition and even in the fields being fetched.
One important point is, whenever you use FOR ALL ENTRIES IN, please make sure that the itab IS NOT INITIAL i.e. the itab must have been filled in. So, place the same conditin before both the SELECT queries like:
IF i_anla[] IS NOT INITIAL.
SELECT bukrs
         anln1
         anln2
         afabe
         gjahr
         peraf
         lnran
         bzdat
         bwasl
         belnr
         buzei
         anbtr
         lnsan
    FROM anep
    INTO TABLE o_anep
    FOR ALL ENTRIES IN i_anla
   WHERE bukrs = i_anla-bukrs
     AND anln1 = i_anla-anln1
     AND anln2 = i_anla-anln2
     AND afabe IN s_afabe
     AND bzdat =< p_date
     AND bwasl IN s_bwasl.
ENDIF.
IF o_anep[] IS NOT INITIAL.
SELECT bukrs
         anln1
         anln2
         gjahr
         lnran
         afabe
         aufwv
         nafal
         safal
         aafal
         erlbt
         aufwl
         nafav
         aafav
         invzv
         invzl
    FROM anea
    INTO TABLE o_anea
    FOR ALL ENTRIES IN o_anep
   WHERE bukrs = o_anep-bukrs
     AND anln1 = o_anep-anln1
     AND anln2 = o_anep-anln2
     AND gjahr = o_anep-gjahr
     AND lnran = o_anep-lnran
     AND afabe = o_anep-afabe.
ENDIF.

Performance wise, a select statement is faster on a view or on a table??

Performance wise, a complex (with multi join) select statement is faster on a view or on a table??

Hi,
the purpose of a view is not to provide performance benefits, it's basically a way to better structure database code and data access. A view is nothing but a stored query. When the optimizer sees references to a view in a query, it tries to merge it (i.e. replace the view with its definition), but it some cases it may be unable to do so (in presence of analytic functions, rownum pseudocolumn etc.) -- in such cases views can lead to a performance degradation.
If you are interested in performance, what you need is a materialized view, which is basically a table built from a query, but then you need to decide how you would refresh it. Please refer to the documentation for details.
Best regards,
Nikolay

Dynamic table name in an inner join - select statement

Hi,
Please can you let me know if is possible to use a Dynamic table name in an inner join?
Something like the statement below? (It works in a simple select statement but not in an inner join)
SELECT *
     INTO CORRESPONDING FIELDS OF <t_itab>
      FROM <Dynamic table name> INNER JOIN pa0050 ON
      ( <Dynamic table name>pernr = pa0050pernr )
       WHERE <Dynamic table name>~pernr = it_pernr-l_pernr
       AND pa0050~bdegr = f_bdegr.
Any help would be apprecited very much.
Thanks & Regards.

Hi,
Check this link.
[http://help.sap.com/saphelp_nw04/helpdata/en/fc/eb39c4358411d1829f0000e829fbfe/frameset.htm]
[Re: accessing dynamic internal table's fields??;
hope it'll help u.
Regards,
Sneha.
Edited by: sneha kumari on Jun 18, 2009 1:57 PM

Adding one more table to the select statement joining 4 tables gives dump

Hi All,
There is a select statement using which four tables namely VBAK,VBAP,LIPS and LIKPUK(view) are joined(inner join).Here, date and time fields are selected from LIPS and used.
My requirement is to consider the Date (LIKP - WADAT_IST) instead of LIPS-ERDAT and time(LIKP-SPE_WAUHR_IST) is to be used instead of LIPS-erzet.
Neither LIPS nor LIKPUK contains time(SPE_WAUHR_IST) field. And, I cannot remove LIPS table or LIKPUK view as each contains a unique field which is used in the report.
When I tried to join LIPS in the select query, it is going to dump.
Can someone suggest a good approach ?
Thanks,
Pavan

Thank you for the detailed explanation of the dump.
The dump details together with your code lead to the answer: 42.
Regards,
Clemens

How to join THREE different tables into internal table using one select statement .

How to join THREE different tables into internal table using one select statement .
Hi experts,
I would like to request your guidance in solving the problem of joining the data from three different database tables into one internal table
Scenario:
Database tables:
SPFLI
SFLIGHT
SBOOK.
Table Fields:
SPFLI - CARRID CONNID COUNTRYFR CITYFRM COUNTRYTO CITYTO
SFLIGHT - CARRID CONNID FLDATE SEATSMAX SEATSOCC SEATSMAX_C
SEATSOCC_C SEATSMAX_F SEATSOCC_F
SBOOK - CARRID CONNID CLASS
MY INTERNAL TABLE IS IT_XX.
Your help much appreciated.
Thanks in advance.
Pawan.

Hi Pawan,
please check below codes. hope it can help you.
TYPES: BEGIN OF ty_xx,
        carrid     TYPE spfli-carrid   ,
        connid     TYPE spfli-connid   ,
        countryfr TYPE spfli-countryfr,
        cityfrom   TYPE spfli-cityfrom ,
        countryto TYPE spfli-countryto,
        cityto     TYPE spfli-cityto   ,
        fldate     TYPE sflight-fldate ,
        seatsmax   TYPE sflight-seatsmax ,
        seatsocc   TYPE sflight-seatsocc ,
        seatsmax_b TYPE sflight-seatsmax_b,
        seatsocc_b TYPE sflight-seatsocc_b,
        seatsmax_f TYPE sflight-seatsmax_f,
        seatsocc_f TYPE sflight-seatsocc_f,
        class      TYPE sbook-class,
      END OF ty_xx,
      t_xx TYPE STANDARD TABLE OF ty_xx.
DATA: it_xx TYPE t_xx.
SELECT spfli~carrid
       spfli~connid
       spfli~countryfr
       spfli~cityfrom
       spfli~countryto
       spfli~cityto
       sflight~fldate
       sflight~seatsmax
       sflight~seatsocc
       sflight~seatsmax_b
       sflight~seatsocc_b
       sflight~seatsmax_f
       sflight~seatsocc_f
       sbook~class
INTO TABLE it_xx
FROM spfli INNER JOIN sflight
ON spfli~carrid = sflight~carrid
AND spfli~connid = sflight~connid
INNER JOIN sbook
ON spfli~carrid = sbook~carrid
AND spfli~connid = sbook~connid.
Thanks,
Yawa

Joining select statements-performance problem

Here two cases are there.Is it posiible to combine the two cases and make it one? I am using OR in the select statement, performance is very bad, i want to improve it.Is there any chance of avoiding that...
1.
select vbeln from vbfa into itab-vbeln where vbelv = final_data-xblnr1
or vbelv = final_data-xblnr2.
select single vbeln from vbrk into tabvbrk-vbeln
where vbeln = itab-vbeln and fkart = 'RTS'.
if sy-subrc eq 0.
move tabvbrk-vbeln to final_data-vbeln1.
modify final_data.
endif.
endselect.
2.
select vbeln from vbfa into itab-vbeln where vbelv = final_data-xblnr1
or vbelv = final_data-xblnr2.
select single vbeln from vbrk into tabvbrk-vbeln
where vbeln = itab-vbeln and fkart = 'ZTR'.
if sy-subrc eq 0.
move tabvbrk-vbeln to final_data-vbeln2.
modify final_data.
endif.
endselect.
Thanks,
fractal

The first main thing is dont use select... endselect.
select vbeln from vbfa into itab-vbeln where vbelv = final_data-xblnr1
or vbelv = final_data-xblnr2.
U can use
select vbeln from vbfa into
corresponding fields of table itab
for all entries in final_data
where vbelv = final_data-xblnr1
or   vbelv = final_data-xblnr2.
Instead of this
select single vbeln from vbrk into tabvbrk-vbeln
where vbeln = itab-vbeln and fkart = 'RTS'.
if sy-subrc eq 0.
move tabvbrk-vbeln to final_data-vbeln1.
modify final_data.
endif.
endselect.
Use
select vbeln from vbrk
        appending table final_data
        where vbeln = itab-vbeln
        and fkart = 'RTS'.
if sy-subrc eq 0.
endif.
Try like this.
Hope this helps.
Kindly reward points for the answer which helped u and helped to solve the problem.

Is it Possible to join 4 or more tables in a Select Statement?

Hello guys,
Can I do Inner Join 4 or 5 tables in a select statement?

U HAVE TO SERCH FOR KEY FIELD FOR THAT . MEANS   HOW TABELS ARE RELATED EACH OTHE FOR THIS U ALSO USE SQ02
FOR EXAMPLE
SELECT        bkpf~belnr
                bkpf~gjahr
                bkpf~bukrs
                bkpf~blart
                bkpf~bldat
                bkpf~budat
                bkpf~usnam
                bkpf~tcode
                bkpf~xblnr
                bsik~lifnr
                bsik~bupla
                bsik~secco
          INTO CORRESPONDING FIELDS OF TABLE itab_pur
          FROM bkpf INNER JOIN bsik
          ON   bkpfbelnr = bsikbelnr
           and bkpfbukrs = bsikbukrs
           WHERE    bkpf~bukrs IN co_code
           AND      bkpf~budat IN pos_dt
           AND      bkpf~tcode IN t_code
           AND      bsik~secco IN se_code.
U CAN ALSO ADD   MORE TABLE BASED ON CONDITION

Fields sequence in the select statement

experts, help needed
itab has following fields,
vbeln posnr erdat erzet vkorg vtweg
kunnr posnr matnr bmeng status.
i have written like this
SELECT avbeln aerdat aerzet avkorg a~vtweg
akunnr bposnr b~matnr
FROM vbak AS a INNER JOIN vbap AS b
ON avbeln = bvbeln
INTO TABLE itab1
FOR ALL ENTRIES IN itab
WHERE a~vbeln = itab-vbeln.
endif.
but my internal table fields sequence are different. so i changed the posnr field in the select query to the second place.
SELECT avbeln bposnr aerdat aerzet avkorg avtweg
akunnr bmatnr
FROM vbak AS a INNER JOIN vbap AS b
ON avbeln = bvbeln
INTO TABLE itab1
FOR ALL ENTRIES IN itab
WHERE a~vbeln = itab-vbeln.
endif.
so in above two select statements which one is right , according to coding standards,?
in this case do i need to use the into corresponding fields addition because i am not selecting bmeng , status fields into the itab?

Hi,
Of the two select statements,
SELECT a~vbeln b~posnr a~erdat a~erzet a~vkorg a~vtweg
a~kunnr b~matnr
FROM vbak AS a INNER JOIN vbap AS b
ON a~vbeln = b~vbeln
INTO TABLE itab1
FOR ALL ENTRIES IN itab
WHERE a~vbeln = itab-vbeln.
is the better one. I,e specifyiig the field sequence exactly as you have in the internal table. Even according to standards, this is better.
If you are chaning the sequence of select fields, then you have to use the INTO CORRESPONDING FILEDS OF TABLE addition, so that the system itself finds the relevant order of fields and fill them automatically.
This is a additional work which has to be performed by the compiler.
Thus its better to use INTO CORRESPONDING FILEDS OF TABLE only when you are using select * , or when you are not sure of the field sequence inthe table.
Regards,
Vik

"Check Statistics" in the Performance tab. How to see SELECT statement?

Hi,
In a previous mail on SDN, it was explained (see below) that the "Check Statistics" in the Performance tab, under Manage in the context of a cube, executes the SELECT stament below.
Would you happen to know how to see the SELECT statements that the "Check Statistics" command executes as mentioned in the posting below?
Thanks
====================================
When you hit the Check Statistics tab, it isn't just the fact tables that are checked, but also all master data tables for all the InfoObjects (characteristics) that are in the cubes dimensions.
Checking nbr of rows inserted, last analyzed dates, etc.
SELECT
T.TABLE_NAME, M.PARTITION_NAME, TO_CHAR (T.LAST_ANALYZED, 'YYYYMMDDHH24MISS'), T.NUM_ROWS,
M.INSERTS, M.UPDATES, M.DELETES, M.TRUNCATED
FROM
USER_TABLES T LEFT OUTER JOIN USER_TAB_MODIFICATIONS M ON T.TABLE_NAME = M.TABLE_NAME
WHERE
T.TABLE_NAME = '/BI0/PWBS_ELEMT' AND M.PARTITION_NAME IS NULL
When you Refresh the stats, all the tables that need stats refreshed, are refreshed again. SInce InfoCube queries access the various master data tables in quereis, it makes sense that SAP would check their status.
In looking at some of the results in 7.0, I'm not sure that the 30 day check is being doen as it was in 3.5. This is one area SAP retooled quite a bit.
Yellow only indicates that there could be a problem. You could have stale DB stats on a table, but if they don't cause the DB optimizer to choose a poor execution plan, then it has no impact.
Good DB stats are vital to query performance and old stats could be responsible for poor performance. I'm just syaing that the Statistics check yellow light status is not a definitive indicator.
If your DBA has BRCONNECT running daily, you really should not have to worry about stats collection on the BW side except in cases immediately after large loads /deletes, and the nightly BRCONNECT hasn't run.
BRCONNECT should produce a lof every time it runs showing you all the tables that it determeined should have stats refreshed. That might be worth a review. It should be running daily. If it is not being run, then you need to look at running stats collection from the BW side, either in Process Chains or via InfoCube automatisms.
Best bet is to use ST04 to get Explain Plans of a poor running InfoCube query and then it can be reviewed to see where the time is being spent and whether stats ate a culprit.

Hi,
Thanks, this is what I came up with:
st05,
check SQL Trace, Activate Trace
Now, in Rsa1
on Cube, Cube1,
Manage, Performance tab, Check Statistics
Again, back to st05
Deactivate Trace
then click on Displace Trace
Now, in the trace display, after scanning through the output,
how do I see the SELECT statements that the "Check Statistics" command executes
I will appreciate your help.

Is it possible to use a case statement when joining different tables based on input parameters?

Hi,
I have a scenario where my stored procedure takes 5 parameters and the users can pass NULL or some value to these parameters and based on the parameters, I need to pull data from various tables.
Is it possible to use a case statement in the join, similar the one in the below example. I'm getting error when I use the below type of statement.
select a.*
from a
case
when parameter1=1 then
inner join a on a.id = b.id
when parameter1=2 then
inner join a on a.id = c.id
end;
Please let me know, if this type of statement works, and if it works will it create any performance issues?. If the above doesn't work, could you please give me some alternate solutions?
Thanks.

Here's a technique for joining A to B or C depending on the input parameters. In theory, you are joining to both tables but the execution plan includes filters to skip whichever join is not appropriate. The drawback is that you have to do outer joins, not inner ones.
CREATE TABLE A AS SELECT LEVEL ak FROM dual CONNECT BY LEVEL <= 100;
CREATE TABLE b AS SELECT ak, bk
FROM A, (SELECT LEVEL bk FROM dual CONNECT BY LEVEL <= 10);
CREATE TABLE c(ak, ck) AS SELECT ak, bk*10 FROM b;
variable p1 NUMBER;
variable p2 NUMBER;
exec :p1 := 1;
exec :p2 := 20;
SELECT /*+ gather_plan_statistics */ A.ak, nvl(b.bk, c.ck) otherk FROM A
LEFT JOIN b ON A.ak = b.ak AND :p1 IS NOT NULL AND b.bk = :p1
LEFT JOIN c ON A.ak = c.ak AND :p1 is null and :p2 IS NOT NULL and c.ck = :p2
WHERE A.ak <= 9;
SELECT * FROM TABLE(dbms_xplan.display_cursor(NULL,NULL,'IOSTATS LAST'));
| Id | Operation             | Name            | Starts | E-Rows | A-Rows |   A-Time   | Buffers |
|   0 | SELECT STATEMENT      |                 |      1 |        |      9 |00:00:00.01 |       7 |
|* 1 | HASH JOIN OUTER      |                 |      1 |      9 |      9 |00:00:00.01 |       7 |
|* 2 |   HASH JOIN OUTER     |                 |      1 |      9 |      9 |00:00:00.01 |       7 |
|* 3 |    TABLE ACCESS FULL | A               |      1 |      9 |      9 |00:00:00.01 |       3 |
|   4 |    VIEW               | VW_DCL_5532A50F |      1 |      9 |      9 |00:00:00.01 |       4 |
|* 5 |     FILTER            |                 |      1 |        |      9 |00:00:00.01 |       4 |
|* 6 |      TABLE ACCESS FULL| B               |      1 |      9 |      9 |00:00:00.01 |       4 |
|   7 |   VIEW                | VW_DCL_5532A50F |      1 |      9 |      0 |00:00:00.01 |       0 |
|* 8 |    FILTER             |                 |      1 |        |      0 |00:00:00.01 |       0 |
|* 9 |     TABLE ACCESS FULL | C               |      0 |      9 |      0 |00:00:00.01 |       0 |
Predicate Information (identified by operation id):
   1 - access("A"."AK"="ITEM_0")
   2 - access("A"."AK"="ITEM_1")
   3 - filter("A"."AK"<=9)
5 - filter(:P1 IS NOT NULL)
   6 - filter(("B"."AK"<=9 AND "B"."BK"=:P1))
   8 - filter((:P2 IS NOT NULL AND :P1 IS NULL))
   9 - filter(("C"."AK"<=9 AND "C"."CK"=:P2))
You can see that table C was not really accessed: the buffer count is 0.
exec :p1 := NULL;
SELECT /*+ gather_plan_statistics */ A.ak, nvl(b.bk, c.ck) otherk FROM A
LEFT JOIN b ON A.ak = b.ak AND :p1 IS NOT NULL AND b.bk = :p1
LEFT JOIN c ON A.ak = c.ak AND :p1 is null and :p2 IS NOT NULL and c.ck = :p2
WHERE A.ak <= 9;
SELECT * FROM TABLE(dbms_xplan.display_cursor(NULL,NULL,'IOSTATS LAST'));
Now table B is not accessed.
| Id | Operation             | Name            | Starts | E-Rows | A-Rows |   A-Time   | Buffers | Reads |
|   0 | SELECT STATEMENT      |                 |      1 |        |      9 |00:00:00.02 |       7 |      2 |
|* 1 | HASH JOIN OUTER      |                 |      1 |      9 |      9 |00:00:00.02 |       7 |      2 |
|* 2 |   HASH JOIN OUTER     |                 |      1 |      9 |      9 |00:00:00.01 |       3 |      0 |
|* 3 |    TABLE ACCESS FULL | A               |      1 |      9 |      9 |00:00:00.01 |       3 |      0 |
|   4 |    VIEW               | VW_DCL_5532A50F |      1 |      9 |      0 |00:00:00.01 |       0 |      0 |
|* 5 |     FILTER            |                 |      1 |        |      0 |00:00:00.01 |       0 |      0 |
|* 6 |      TABLE ACCESS FULL| B               |      0 |      9 |      0 |00:00:00.01 |       0 |      0 |
|   7 |   VIEW                | VW_DCL_5532A50F |      1 |      9 |      9 |00:00:00.01 |       4 |      2 |
|* 8 |    FILTER             |                 |      1 |        |      9 |00:00:00.01 |       4 |      2 |
|* 9 |     TABLE ACCESS FULL | C               |      1 |      9 |      9 |00:00:00.01 |       4 |      2 |

Select statement Table Join sequence (Performance)

Similar Messages

Maybe you are looking for