Quantcast
Channel: SCN : All Content - ABAP Development
Viewing all articles
Browse latest Browse all 8332

Performance issue when using For all entries with Inner Join on Large tables

$
0
0

Hi ,

 

I am facing on issue of bad performance when trying to query AUSP table using For all entries and Inner join on

INOB and AUSP table .

 

Below are the 2 queries , I am using :

 

1) First query is on AENR table , to fetch the Change masters(AENNR field) , which are around 1 , 90 ,000 records  .  This works well with no performance issue .

 

2) In the second query ,I am trying to fetch the Change master characteristic data(ATWRT field) for all the change masters fetched in the first query , from AUSP table with a join on INOB table ,  for a given characteristic(ATINN) value .

But I am facing a big performance issue here .

 

Below is the SQL query execution plan for the query .

SELECT

  T_00."OBJEK", T_01."ATWRT"

FROM

  "INOB" T_00 INNER JOIN "AUSP" T_01 ON T_01."MANDT"=:A0 AND T_00."CUOBJ"=T_01."OBJEK" AND T_00."KLA

  RT"=T_01."KLART"

WHERE

  T_00."MANDT"=:A1 AND T_00."OBTAB"=:A2 AND T_00."OBJEK" IN (:A3 ,:A4 ,:A5 ,:A6 ,:A7 ,:A8 ,:A9

  ,:A10 ,:A11 ,:A12 ,:A13 ,:A14 ,:A15 ,:A16 ,:A17 ,:A18 ,:A19 ,:A20 ,:A21 ,:A22 ,:A23 ,:A24 ,:A25

  ,:A26 ,:A27 ,:A28 ,:A29 ,:A30 ,:A31 ,:A32 ,:A33 ,:A34 ,:A35 ,:A36 ,:A37 ) AND T_00."KLART"=:A38

  AND T_01."KLART"=:A39 AND T_01."ATINN"=:A40 AND T_01."ATWRT"=:A41

 

 

Execution Plan

 

 

 

 

SELECT STATEMENT ( Estimated Costs = 2 , Estimated #Rows = 1 )

 

        6 FILTER

          Filter Predicates

 

            5 NESTED LOOPS

 

                3 NESTED LOOPS

                  ( Estim. Costs = 1 , Estim. #Rows = 1 )

                  Estim. CPU-Costs = 9,061 Estim. IO-Costs = 1

 

                    1 INDEX RANGE SCAN AUSP~N1

                      ( Estim. Costs = 1 , Estim. #Rows = 1 )

                      Search Columns: 4

                      Estim. CPU-Costs = 5,747 Estim. IO-Costs = 1

                      Access Predicates Filter Predicates

                    2 INDEX UNIQUE SCAN INOB~0

                      Search Columns: 2

                     Estim. CPU-Costs = 1,804 Estim. IO-Costs = 0

                     Access Predicates

 

               4 TABLE ACCESS BY INDEX ROWID INOB

                 Estim. CPU-Costs = 3,314 Estim. IO-Costs = 0

                 Filter Predicates

 

Please provide any comments or suggestions

 

Thanks


Viewing all articles
Browse latest Browse all 8332

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>