on 04-06-2009 12:11 PM
Good morning
We have a problem in a SAP 46C productive during the access to table
CE4TLCI (CE4XXXX) in program
RK2ATLCI_POST (RK2Axxxx_POST). We have a massive selection of the same
table ending in TIME_OUT.
I've readen some notes about this long access , above all 551750 and
35288. These notes
describes the correct index to define obtaining a better access on
table .
Anyway the access seems good using the secondary index that i show you:
I think the select is the problem but i don't know how to increase the
speed...also the buffer gets is very high.
SELECT
/*+
FIRST_ROWS(5)
*/
FROM
"CE4TLCI"
WHERE
"MANDT" = :A0 AND "AKTBO" = :A1 AND "KNDNR" = :A2 AND "ARTNR" = :A3 AND "FKART" = :A4 AND "KAUFN"
= :A5 AND "KDPOS" = :A6 AND "AUFNR" = :A7 AND "BUKRS" = :A8 AND "KOKRS" = :A9 AND "WERKS" = :A10
AND "GSBER" = :A11 AND "VKORG" = :A12 AND "VTWEG" = :A13 AND "SPART" = :A14 AND "PRCTR" = :A15
AND "PPRCTR" = :A16 AND "KSTRG" = :A17 AND "PSPNR" = :A18 AND "MEINH" = :A19 AND "KDGRP" = :A20
AND "COPA_KOSTL" = :A21 AND "BRSCH" = :A22 AND "WWASA" = :A23 AND "WWSPS" = :A24 AND "GEBIE" =
:A25 AND "WWBE" = :A26 AND "WWDTC" = :A27 AND "VKBUR" = :A28 AND "WWVC1" = :A29 AND "WWVC2" =
:A30 AND "WWVC3" = :A31 AND "WWVC4" = :A32 AND "WWVC5" = :A33 AND "WWVC6" = :A34 AND "WWVC7" =
:A35 AND "WWVC8" = :A36 AND "WWVC9" = :A37 AND "WWVDC" = :A38 AND "VV005_ME" = :A39 AND
"VV006_ME" = :A40 AND "VV007_ME" = :A41 AND "VV008_ME" = :A42 AND "VV009_ME" = :A43 AND
"ABSMG_ME" = :A44 AND "VV001_ME" = :A45 AND "VV002_ME" = :A46 AND "VVCON_ME" = :A47 AND
"VV03A_ME" = :A48 AND "VVIMP_ME" = :A49 AND "VVQNT_ME" = :A50 AND "VVQSD_ME" = :A51 AND
"VVDCD_ME" = :A52 AND "VVTES_ME" = :A53 AND ROWNUM <= :A54#
Execution Plan
SELECT STATEMENT ( Estimated Costs = 2 , Estimated #Rows = 1 )
5 COUNT STOPKEY
5 TABLE ACCESS BY INDEX ROWID CE4TLCI
INDEX RANGE SCAN CE4TLCI~1
The index CE4TLCI~1 is so structured.....
MANDT 1
AKTBO 1
KNDNR 5.442
ARTNR 6.922
BUKRS 5
WERKS 78
VTWEG 22
Last statistics date 02.04.2009
Analyze Method Sample 214.681 Rows
Levels of B-Tree 3
Number of leaf blocks 81.689
Number of distinct keys 764.356
Average leaf blocks per key 1
Average data blocks per key 3
Clustering factor 2.347.350
How can i try to getting better performance to this select?
Thanks in advance
Nicola
in addition:
How big is the table?
what is your oracle version (10g, but 10202 or 10204)?
have you installed all recommended patches?
what optimizer merge fix patch is installed?
The index is goodish, it has 7 fields out of 54.
the problem (probably) is that the fields do not filter many rows and a lot of accesses have to be done to retrieve the 5 rows you are looking for.
This is difficult to say without knowing
the amount of rows in your table
2 million?
20 million?
the selectivity of the index fields
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hey there,
you may want to use the script of note
1257075 SQL_ID Data Collector (Oracle >= 10.2.0.2)
to find out more about the runtime of this statement.
Also:
Make sure the DB parameters are set as recommended - be aware: a high buffer cache hit ratio does not mean a good response time at all.
regards,
Lars
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
85 | |
10 | |
10 | |
9 | |
6 | |
6 | |
6 | |
5 | |
4 | |
3 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.