Use plan from stored_outlines from query with different syntax?

Hello,
I'd like to know is there any ability to use via stored_outlines plan from query with different syntax?
Database version - 11.1.06
If yes, maybe you'll take a look at steps (based on Metalink 726802.1), what I tried to do to make it work (with no success):
alter session set create_stored_outlines=good;
run good query
create private outline good from SYS_OUTLINE_0... -< GOOD
alter session set create_stored_outlines=bad;
run bad query
create private outline good from SYS_OUTLINE_0... -< BAD
update ol$ set sql_text, textlen, signature, hash_value, hash_value2 from BAD to GOOD records
leave ol$hintcount for GOOD untouched
delete ol$ where ol_name='BAD';
delete ol$hints where ol_name='BAD';
delete ol$nodes where ol_name='BAD';
execute dbms_outln_edit.refresh_private_outline('GOOD');
alter session set sql_trace=true;
alter session set use_private_outlines=true;
run bad query
examine trace file and find that it use execution plan different from both bad and good queries
PS: Originally bad query is posted already Re: Poor performance while join of 2 comprehensive large views and small table
PPS: Also I review thread CBO not picking correct indexes or doing Full Scans
Thanks,
Sergiy
Edited by: kiberpress on Sep 30, 2009 6:59 AM

A query with different syntax would result in a different hash value.
Stored outlines work based on hash value.
Your question is probably answered now.
Sybrand Bakker
Senior Oracle DBA

Similar Messages

Same query with different explain plans

Hi All,
I have one of the select query with different explain plans on two separate env, the query fetches the correct index on test env whereas on prod it's not fetching the same index.
The structure, indices, no. of rows are similar in CRMINFO table with up-to-date statistics.
Env Details :
OS - Sun Solaris 5.10
DB - 10.2.0.4
Init param:
Optimizer_mode = ALL_ROWS
optimizer_dynamic_sampling integer 5
optimizer_features_enable string 10.2.0.4
optimizer_index_caching integer 90
optimizer_index_cost_adj integer 30
Query :*
SELECT COUNT (*)
FROM CRMINFO
WHERE RETAILER = :1
AND STATUS = 20
AND EXC = :1
AND SUBNO IS NULL
Expain Plan (TST):
SELECT STATEMENT ALL_ROWSCost: 916 Bytes: 19 Cardinality: 1
3 SORT AGGREGATE Bytes: 19 Cardinality: 1
2 TABLE ACCESS BY INDEX ROWID TABLE TST.CRMINFO Cost: 916 Bytes: 16,663 Cardinality: 877
1 INDEX RANGE SCAN INDEX TST.CRMINFO_X1 Cost: 42 Cardinality: 12,549
Index (TST):
CRMINFO_X1(EXC, RETAILER, STATUS)
Explain Plan (PROD):
SELECT STATEMENT ALL_ROWSCost: 1,832 Bytes: 19 Cardinality: 1
3 SORT AGGREGATE Bytes: 19 Cardinality: 1
2 TABLE ACCESS BY INDEX ROWID TABLE PROD.CRMINFO Cost: 1,832 Bytes: 2,052 Cardinality: 108
1 INDEX RANGE SCAN INDEX PROD.CRMINFO_X2 Cost: 117 Cardinality: 42,519
Index (PROD):
CRMINFO_X2 (RETAILER)
How does Oracle calculates the cost and decides which index it should fetch ? Why it didn't choose the same index as of test env? How should i approach and which domains i need to dig-in to find the cause?
I did try playing with the above mentioned init parameters but it didn't help at all.
Thanks.
Regards,
~Pointer

Pointer wrote:
Hmm, my worry is how do i force oracle to grap the proper index on prod i.e (CRMINFO_X1). I certainly believe it's a bad approach on adding a hint in the select statement and to force oralce to fetch that index.Why do you believe that, the index you mention is the "proper" index versus what Oracle is choosing? Can you prove with hinting that the "proper" index results in a faster and more efficient execution plan? If it does, then the next place I would look at is the statistics for the tables and columns of interest. From here you could try and estimate why Oracle thinks the other index is better. Another option is to run a 10053 (CBO) trace and see why Oracle thinks it is better.
I would not support a hint in a production environment, except in the most extreme cases. Usually the CBO makes the right choice, but it only can if the statistics match the distribution of data.
Refreshing the data may help me simulating the issue on TST but it wouldn't help me to understand why on prod it uses CRMINFO_X2 instead of CRMINFO_X1 which has all the three columns in the Where clause of the query.It would help because it's a test environment and you wouldn't have to do any queries directly on your production system to achieve the same results.
>
A bad thought here :( , if i create a new index by changing the column positioning say like ( RETAILER, STATUS , EXC) instead of (EXC, RETAILER, STATUS) will oracle fetch it ? or would it help in reducing the cost and cardinatlity of the select query.It's not that easy. There is a lot that goes into the cost calculation, some of that is known (through the great work by Jonathan Lewis and Richard Foote), and some of that is purely internal to Oracle. If you are more interested in the internals Cost-Based Oracle Fundamentals by Jonathan Lewis is a great book.
HTH!

Can I re-use a parameterized manual test case with different set of parameters in a different test plan?

Can I re-use a parameterized manual test case with different set of parameters in a different test plan without impacting the parameters in the first test plan?

Hi ssteele_1,
Thank you for posting in MSDN forum.
Generally, I know that if you want to re-use a parameterized manual test case and without impacting the parameters in the first test plan.
I suggest you could try to right click this parameterized manual test case ->Select the Create copy and add to suite option like the following screen shot.
Then the parameterized manual test case can be re-used as a new test case in this MTM and it will did not impact the original parameterized manual test case.
Best Regards,
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

How can i transfer my contacts from iphone4s to imac with different apple id without loosing everything on my phone

how can i transfer my contacts from iphone4s to imac with different apple id without loosing everything on my phone

Move the iPhoto library from the Pictures item of the Finder's sidebar on the MacBook Pro to the same location on the iMac via FireWire Target Disk mode, or a drive, partition, or disk image formatted as Mac OS Extended, or through AFP file sharing over a network.
(68308)

Simple spatial query with different SRID

Can anyone help me with just a real simple spatial query with different SRID.
Here is the error i get when i perform the query:
AND SDO_RELATE(A.geometrie_point, B.GEOMETRIE_POLYGONE, 'mask=anyinteract querytype=WINDOW') = 'TRUE'
ERROR at line 4:
ORA-00604: error occurred at recursive SQL level 1
ORA-13207: incorrect use of the [IN COMPATIBLE BOUNDS in SDO_RELATE] operator
ORA-06512: at "MDSYS.SDO_INDEX_METHOD", line 84
ORA-06512: at line 4

Hi,
I am using 8.1.7.2 and i have the lattest spatial patch.
From the start i had different SRID. The only thing i changed was the bounds of the coordinate system.
I reset my bounds of the coordinate system with SRID 8307 to -180,
180 in X (longitude), and -90,90 in Y (latitude) to see if the query window geometry
(geom2) will be transformed to the coordinate system of the layer geometries (geom1)
Her is the content of the user_sdo_geom_metadata :
GIS_PCP GEOMETRIE_POINT
SDO_DIM_ARRAY(SDO_DIM_ELEMENT('X', -5800000, 3600000, .00000005), SDO_DIM_ELEMENT('Y', 3516000, 6000000, .00000005))
82227
GIS_TEST GEOMETRIE_POLYGONE
SDO_DIM_ARRAY(SDO_DIM_ELEMENT('X', -180, 180, .00000005), SDO_DIM_ELEMENT('Y', -90, 90, .00000005))
8307
Here is my query :
select /*+ ordered */ A.no_point
from gis_pcp A, gis_test B
WHERE SDO_RELATE(A.geometrie_POINT, B.GEOMETRIE_POLYGONE, 'mask=anyinteract querytype=WINDOW') = 'TRUE';
Here is my result :
WHERE SDO_RELATE(A.geometrie_POINT, B.GEOMETRIE_POLYGONE, 'mask=anyinteract querytype=WINDOW') = 'TRUE'
ERROR at line 3:
ORA-00604: error occurred at recursive SQL level 1
ORA-13207: incorrect use of the [IN COMPATIBLE BOUNDS in SDO_RELATE] operator
ORA-06512: at "MDSYS.SDO_INDEX_METHOD", line 84
ORA-06512: at line 4
If i change the bounds of GIS_TEST to the exact bouns of GIS_PCP then i dont have the error message above and i have correct results. Do i have the adjust the bounds of every layer to fit the biggest bound?

Why can't two different users use Itunes on the same computer with different log ins

Why can't two different users use Itunes on the same computer with different log ins

Pmorgan5672 wrote:
Why can't two different users use Itunes on the same computer with different log ins
They can, but not at the same time.
If one user left iTunes running, and another user tries to use iTunes from a different Windows user ID, that second user will get an error message. If they really want to use iTunes, they either have to beg the first user to log in and close iTunes, or else they have to restart the computer.
If you are asking the technical reason why, it is something about context switching.

Can I use iCloud to sync two devices with different apples Ids?

Can I use iCloud to sync two devices with different apple ID's

What do you mean by "apple ID"? There's an ID you can use at the iTunes store to buy songs, apps, etc. there's an ID you can use for your iCloud account - the two can be different or the same. So which service are you referring to when you talk about an apple ID?

Query with 'different from' clause in Sqlite

My question is very simple:
How can I submit a query with a 'different from' clause into a table.
In AIR, I would like to query :
"select * from myTable where myField <> 'value' ";
but it does'nt work !
Nor
"select * from myTable where myField != 'value' ";
nor
"select * from myTable where myField is not 'value' ";
is it impossible to query with a 'different from' clause in Sqlite wih Adobe Air

Hello,
SQLite in AIR does support 'different from' and '<>' is the correct operator. Take the following emp table for example:
id
name
age
1
A
26
2
B
30
3
C
28
If you want to query the records whose name isn't A. The following SQL is correct:
select * from emp where name <> 'A'
Thanks,
Yang

Repeated rows resulting from executeWithParams on query with bind variable

hi
Please consider this example application created using JDeveloper 11.1.1.3.0
at http://www.consideringred.com/files/oracle/2010/RepeatedRowsIssueApp-v0.01.zip
It has a View Object "EmployeesVO" based on a SQL query with a bind variable "EmailBVar".
The page "findEmployees.jspx" has executeWithParams dropped as an "ADF Parameter Form..." and the "EmployeesVOVI" collection as a table.
On "EmployeesVOVIIterator" the RangeSize has been modified from the default 25 to 10.
So, all straightforward stuff.
A scenario (sc1), running the page "findEmployees.jspx", specifying a value and clicking the "ExecuteWithParams" button results in repeated rows shown in the table, that is the first row (and subsequent rows) are repeated at row 11 (and subsequent rows),
as can be seen in the screencast at http://screencast.com/t/bn9QvV0qt
question:
- (q1) How can the repeated rows in scenario (sc1) be explained/avoided?
many thanks
Jan Vervecken

Thanks for your reply Nick Haralabidis.
it can be avoided by setting the EmployeesVO Access Mode to Range Paging and setting the Range Size to 10. Indeed, that does seem to avoid the repeating rows.
Now why you get this behavior when you don't specify Range Paging on the VO I don't know. It could be because: 1) it is a bug - the obvious answer ;) , 2) because setting the RangeSize on the EmployeesVOVIIterator without setting the Access Mode on the EmployeesVO to Range Paging is conflicting - but then again there is no warning or error produced by JDeveloper, 3) ...It does look like unintended behaviour, for a "default approach" like this.
In documentation section "39.1.5 Efficiently Scrolling Through Large Result Sets Using Range Paging " ...
at http://download.oracle.com/docs/cd/E14571_01/web.1111/b31974/bcadvvo.htm#ADFFD1179
... this Range Paging for a View Object seems to be described as an optimization, not as a requirement for functionally correct behaviour.
- about (q1)
Another approach that seems to avoid the repeated rows in scenario (sc1) is removing the value for the RangeSize attribute on the iterator, so the RangeSize is not specified.
Using the modified example application
at http://www.consideringred.com/files/oracle/2010/RepeatedRowsIssueApp-v0.02.zip
... this can be seen in scenario (sc2) in the screencast at http://screencast.com/t/0bpCo33i8J
Although, the documentation in section "22.4.2.2 Iterator RangeSize Attribute " ...
at http://download.oracle.com/docs/cd/E14571_01/web.1111/b31974/web_form.htm#ADFFD706
... does not seem to mention anything about not specifying the RangeSize attribute, it does say "... You can set it to -1 to have the full record set returned. ...".
As can be seen in scenario (sc2) the EL expression #{bindings.EmployeesVOVIIterator.rangeSize} resolves to -1 which suggests that not specifying the RangeSize attribute is the same as configuring the RangeSize attribute to -1.
questions:
- (q2) What is the preferred way to avoid having "the full record set returned", without having repeated rows as in scenario (sc1)?
- (q3) Is the default configuration of RangeSize="25" on the iterator intended behaviour (for an interator supporting an executeWithParams action)?
regards
Jan

Query with different parameter take different time to execute.

Hi,
I am a C/C++ programmer and newbie to database. I find weird case to my company database. oracle 10g.
I have a query (below). When I set GENRE_ID value to 20, query execution time only take *5* seconds. But when I change to 10, it take *54* seconds! The same query but different values GENRE_ID (20 or 10) -- it also happen to other GENRE_ID values.
on below query
WHERE substr(GENRE_ID,0,2) = 10 and GENRE_ID != 10) take 54 seconds
WHERE substr(GENRE_ID,0,2) = 20 and GENRE_ID != 20) take 5 seconds
The explain plan give exact same result for both queries.
I have follow Re: 3. How to improve the performance of my query? / My query is running slow. for optimizer but the problem still there.
And we have rebuilt all indexes.
Anyone can help me? I need any advises for this problem.
Note:
Song table only have 570K records and Song_vod table 1100 records.
Query:
select *
from(
select rownum rowno, a.*
from(
select a.SONG_VOD_ID, a.SONG_ID, a.VOD_TITLE, a.ALBUM_ID, a.ARTIST_ID, a.VOD_ISSUE_DATE, a.VOD_ORDER_ISSUE_DATE, a.VOD_ADULT_YN, a.VOD_L_IMG_PATH, a.VOD_M_IMG_PATH, a.VOD_S_IMG_PATH, a.RECOMMEND_CNT, a.OPPOSE_CNT, get_song_name(a.SONG_ID) SONG_NAME, get_album_name(a.ALBUM_ID) ALBUM_NAME, get_artist_name(a.ARTIST_ID) ARTIST_NAME, VOD_ISSUE_DATE vodIssueDate
from SONG_VOD a inner join SONG b on a.SONG_ID = b.SONG_ID
where a.LC_STATUS_CD = 'CS0006'
AND (b.GENRE_ID
in (
select GENRE_ID
from MD_GENRE
WHERE substr(GENRE_ID,0,2) = 10
and GENRE_ID != 10)
OR b.GENRE_ID
in (
select GENRE_ID
from MD_GENRE
WHERE substr(GENRE_ID,0,2) = '40'
and GENRE_ID != '40'
) order by a.REG_DATE desc, a.SONG_VOD_ID desc
) a WHERE rownum <= 0 + 30
Thank you,
Regards
-=Rika Chaniago=-

907814 wrote:
I am a C/C++ programmer and newbie to database. I find weird case to my company database. oracle 10g.
I have a query (below). When I set GENRE_ID value to 20, query execution time only take *5* seconds. But when I change to 10, it take *54* seconds! The same query but different values GENRE_ID (20 or 10) -- it also happen to other GENRE_ID values.This is to be expected. Even an IDENTICAL query can (and often will) have DIFFERENT execution times.
The typical reason is how fast the query process can get to the data block(s) containing the relevant row(s). Is that data block still on disk and require expensive and slow physical I/O to move it into the buffer cache for use? Is that data block already cached for much faster logical I/O access? If in memory, is there any contention in getting a latch for the chain the data block hangs off from (how hot is that block)? Etc.
The run-time environment is not static. Thus execution times of queries (called cursors in Oracle) is not static. Execution times are expected to vary.
Some basics. SQL code is source code. SQL code needs to be parsed and converted into a set of instructions that the server can execute. So it is similar to C/C++. SQL source code needs to be compiled. This executable code in Oracle is called a cursor.
Like your code, the cursor can be executed with different input variables (bind variables). However, that code would have been compiled using certain assumptions about the data. And that executable code may work fine for some input data, and work not so okay for other input data. As the input parameter values are not not equal. E.g. there are a 1000 rows to process when employee type parameter is "employee" and only 10 rows when it is "manager". The code could have been compiled with the assumption that 10 rows would be the average for all input parameters - the Cost Based Optimiser needs to base its decision on the best execution path for that SQL source code on certain assumptions about the data. These may not always be correct. (usually due incorrect or stale stats about the data)
Thus you also need to look at what the execution plan is (the URL for which has already been supplied).
The Oracle® Database Performance Tuning Guide is at http://www.oracle.com/pls/db112/portal.all_books

Query with different currency for different users

Hi at all,
just a crazy thing happened. I created a query. Every user gets the result with the currency EURO.
Only one user here gets the country currency like US$, Francs etc.
If he logs in on a different PC, he also gets EURO.
So I think the problem is because of his PC.
I already checked his user data in SAP and BW.
And he really uses the same query with the same filters...
Thanks for helping...
Kind regards from totally confused Marielle

Hi,
Check the Front end Patch levels on both the machines. If they are different, then you can decide what you want to do, either upgrade it on other PCs or not use the one that has it.
Cheers,
Kedar

Huge difference in Execution time in same Query with different Parameter

Hi Experts,
We are facing an unique problem once we are executing the query in HANA SQL prompt. This Query was generated from BObj and executing in HANA system. Once this query running with following condition, it is taking almost 7-00 minute to execute and returning around 924 rows.
<< WHERE
Table__1."LOGSYS" IN ('RKGCLNT102')
AND
Table__1."CompanyCode" IN ('7240','7245')
AND
Table__1."Plant" IN ……………… >
However if we run the same query with some different plant, It is taking only 2 second. Please find the Query here.
<< WHERE
Table__1."LOGSYS" IN ('RKGCLNT102')
AND
Table__1."CompanyCode" IN ('7245','7600')
AND
Table__1."Plant" IN ……………… >
This is really an unexpected behavior and we are not able to get the actual reason why it is behaving like this.
Could anyone please help to analyze this issue to fine the root cause.
Thanks in Advance.
Regards
Satrajit.

Hi there
Unfortunately you provided too few information to analyze the issue.
Maybe the underlying tables have very skew data and the first select has to read a larger share of the base tables.
Maybe the columns had been unloaded before and the first query had to load them into memory first.
Is the runtime always bad with the one and always good with the other set of parameters?
Have you checked the PlanViz for both versions? How do these differ?
- Lars

Simulate query with different user

Hello all,
sometimes I receive calls from our user in BW that some query is not running/no authorization, but when i run the query it works, because I have rights to see everything.
Is there any way to simulate how a query is running under a different user (which has restricted authorization). I want to check e.g authorization but I do not want to copy existing user to a test user.
I saw in the TA RSTRACE that I can enetr a different user, but how can I run a query with this user profile.
Any help/ideas would be great. Thank you in advance.
Best Regards,
Stefan from Munich/Germany

Hello Edwin,
works perfect. Thank you.
BR,
Stefan

Is it possible to use parameter in a user query with subquery ?

Hi ,
i need to use parameter ( [%0] ) in user query using a subquery , but i have errors message when i execute the query
example :
SELECT T0.[DocNum] as docnum, T0.[DocTotal] as cde , 0 as fact FROM ORDR T0 WHERE T0.[CardCode] =N'[%0]'
union all
SELECT T0.[DocNum], 0 as cde,T0.[DocTotal] as fact FROM OINV T0 WHERE T0.[CardCode] =N'[%0]'
works fine
but ( exactly the same query in the subquery )
Select t2.Docnum ,t2.cde , t2.fact from
SELECT T0.[DocNum] as docnum, T0.[DocTotal] as cde , 0 as fact FROM ORDR T0 WHERE T0.[CardCode] =N'[%0]'
union all
SELECT T0.[DocNum], 0 as cde,T0.[DocTotal] as fact FROM OINV T0 WHERE T0.[CardCode] =N'[%0]'
) t2
return an error message
the same query without parameter works fine too :
Select Docnum ,cde , fact from
SELECT T0.[DocNum] as docnum, T0.[DocTotal] as cde , 0 as fact FROM ORDR T0 WHERE T0.[CardCode] ='C1'
union all
SELECT T0.[DocNum], 0 as cde,T0.[DocTotal] as fact FROM OINV T0 WHERE T0.[CardCode] ='C1'
) t2
Is it possible to use parameter in a subquery ?
thanks

Try this one:
declare @cc nvarchar(20)
set @cc=/*Select c.cardcode from OCRD c where c.cardcode=*/'[%0]'
Select Docnum ,cde , fact from
SELECT T0.[DocNum] as docnum, T0.[DocTotal] as cde , 0 as fact FROM ORDR T0 WHERE T0.[CardCode] =@cc
union all
SELECT T0.[DocNum], 0 as cde,T0.[DocTotal] as fact FROM OINV T0 WHERE T0.[CardCode] =@cc
) t2

Combine tables from 2 SQL servers with different Schemas and update as new data is entered

I have 2 SQL server 2000 machines, I need to take a table from each one and combine them together based on a date time stamp. The first machine has a database that records information based on an event it is given a timestamp the value of variable is stored and a few other fields are stored in Table A. The second machine Table B has test data entered in a lab scenario. This is a manufacturing facility so the Table A data is recorded by means of a third party software. Whenever a sample is taken in the plant the event for Table A is triggered and recorded in the table. The test data may be entered on that sample in Table B several hours later the lab technician records the time that the sample was taken in Table B but it is not exact to match with the timestamp in Table A. I need to combine each of these tables into a new SQL server 2005 database on a new machine. After combining the tables which I am assuming I can based on a query that looks at the timestamp on both Tables A & B and match the rows up based on the closest timestamp. I need to continuously update these tables with the new data as it comes in. I havent worked with SQL for a couple of years and have looked at several ways to complete this task but havent had much luck. I have researched linked servers, SSIS, etc Any help would be greatly appreciated.

Hi Catalyst,
Welcome to the MSDN Forums!
I would have a concern with respect to your issue. If there is no way to identify a sample, save for the closest timestamp on table A - what would happen if two samples were taken in the plant, and the test results were entered in reverse order? The samples would not tie back to the appropriate measurements. Can't you get an identifier of some sort to tie that sample back to table A?
If you can't, something like this should work, but I'm not sure how you'll "serialize" the samples to avoid the situation above.
Code Snippet
--Create sample data
DECLARE @TableA TABLE ( SampleNo int NOT NULL,
SampleTime datetime);
DECLARE @TableB TABLE ( SampleTakenTime datetime,
SampleColour varchar(50),
SampleQuality char(1),
SampleResultNo int);
INSERT @TableA SELECT 1, GETDATE()
WAITFOR DELAY '00:00:10';
INSERT @TableB SELECT GETDATE(), 'Orange', 'A', '5'
WAITFOR DELAY '00:00:20';
INSERT @TableB SELECT GETDATE(), 'Purple', 'D', '7'
WAITFOR DELAY '00:00:10';
INSERT @TableA SELECT 2, GETDATE()
--Actual Query begins now
SELECT *
FROM @TableA AS a
JOIN @TableB AS b ON ABS(DATEDIFF(millisecond,a.SampleTime,b.SampleTakenTime)) = (SELECT MIN(ABS(DATEDIFF(millisecond,ta.SampleTime,tb.SampleTakenTime)) )
FROM @TableB tb
JOIN @TableA ta ON a.SampleNo = ta.SampleNo)
Now, to account for the fact that you are querying data from a different server, you need to create a linked server (see sp_addlinkedserver and sp_addlinkedsrvlogin) and then use the fully qualified name (server.database.schema.table) to reference the table.
Hope this helps,
Aaron

Use plan from stored_outlines from query with different syntax?

Similar Messages

Maybe you are looking for