Performance Problem in SQL Statement

Hi,
I am using a SELECT statement where the execution time is taking longer time:
SELECT MSD,IMS,BOOK_IND FROM RW_TABLE,TABLE(GET_SUD)
WHERE SERVED_MSISDN=MSD
(MSD Column is a Object type from TABLE Function TABLE(GET_SUD) and (SERVED_MSISDN is a column of the table RW_TABLE)
Any help will be needful for me
Thanks and Regards

Your query:
SELECT MSD,IMS,BOOK_IND
FROM RW_TABLE,TABLE(GET_SUD)
WHERE SERVED_MSISDN=MSDYou reference two tables - RW_TABLE and GET_SUD - yet you have no join between them. Hence Oracle is doing a CARTESIAN PRODUCT between the two data sets. As each table or data set gets bigger, so will the results, and so will the elapsed time.
You need to join the two tables together properly. Read a book on Relational Databases if you do not understand what a join is between tables, and why it is needed.
I would also guess that you lack an index on the SERVED_MSISDN column on its own, or as the leading column in an index. Instead Oracle is reading every entry in an index to find any matching rows (FAST FULL SCAN CTVA_ETL.TEMP_INDX). So again, as the tables get bigger so the FULL SCAN of the whole index will take longer too.
John
[Database Performance Blog|http://databaseperformance.blogspot.com]

Similar Messages

Strage Performance Problems with SQL Statements....

Hi,
I have realized a strange performance collapse while changing simple things in SQL Statements...
I have build a simple Statement:
SELECT
a.AAA,
a.BBB,
a.CCC,
b.AAA
from
TABLE_A a,
TABLE_B b,
TABLE_C c
where
a.XXX=b.XXX
and c.XXX = a.XXX
and c.yyy = 'SOMETHING'
Its very fast, even with complex XSL Transformations...
After Putting a GROUP BY or DISTINCT into the statement (to surpress Dataset "Clones") it tooks around 100 times more Time then before....
I have tested the statement in a SQLPlus, it was as fast as before, but in the XMLPublisher it tooks much longer.....
Have you realizesd this Problem before?
Greetings...

PROBLEM SOLVED !
It has to be 8.1.6 across the whole environment.
null

Performance problem with SQL in NT vs. Solaris

Hi, I am using the thin driver in classes12.zip. The 8.1.5 server runs on a solaris box. With the same code executing the same sql statement against the same database using the same version of JRE, Statement.execute() takes 5-10 times longer on a solaris client then a NT client. The other java code doesn't show any noticeable performance degradation. Has anyone seen this problem before? Thanks in advance.

JDK1.3.0 on NT and JDK1.2.2/JDK1.3.0Beta on Solaris (tried both jdk on solaris, pretty much the same result). The network set up is this:
1 Oracle Server (8.1.5)
1 Web Server (Alliaire JRun, running servlets, communicates with oracle server via JDBC thin driver)
a couple clients using explorer or netscape
The problem is actually the web server. When the webserver is installed on NT, a sql call (stmt.executeUpdate(sql) for example) takes say 10 seconds, the same call will take twice as long or more on Solaris. The reason why I don't think JRun solaris/NT version makes that much of a difference is that besides the JDBC calls, all the other java codes completes at about the same rate.
null

Performance Problem - MS SQL 2K and PreparedStatement

Hi all
I am using MS SQL 2k and used PreparedStatement to retrieve data. There is strange and serious performance problem when the PreparedStatement contains "?" and using PreparedStatement.setX() functions to set its value. I have performed the test with the following code.
for (int i = 0; i < 10; i ++) {
try {
con = DBConnection.getInstance();
statement = con.prepareStatement("SELECT * FROM cardno WHERE car_no = '" + cardNo + "'");
// statement = con.prepareStatement("SELECT * FROM cardno WHERE car_no = ?");
// statement.setString(1, cardNo);
rs = statement.executeQuery();
if (rs.next()) {
catch(SQLException e) {
e.printStackTrace();
finally {
try {
rs.close();
statement.close();
catch(SQLException e) {
e.printStackTrace();
Iteration Time (ms)
1 961
10 1061
200 1803
for (int i = 0; i < 10; i ++) {
try {
con = DBConnection.getInstance();
// statement = con.prepareStatement("SELECT * FROM cardno WHERE car_no = '" + cardNo + "'");
statement = con.prepareStatement("SELECT * FROM cardno WHERE car_no = ?");
statement.setString(1, cardNo);
rs = statement.executeQuery();
if (rs.next()) {
catch(SQLException e) {
e.printStackTrace();
finally {
try {
rs.close();
statement.close();
catch(SQLException e) {
e.printStackTrace();
Iteration Time (ms)
1 1171
10 2754
100 18817
200 36443
The above test is performed with DataDirect JDBC 3.0 driver. The one uses ? and setString functions take much longer to execute, which supposed to be faster because of precompilation of the statement.
I have tried different drivers - the one provided by MS, data direct and Sprinta JDBC drivers but all suffer the same problem in different extent. So, I am wondering if MS SQL doesn't support for precompiled statement and no matter what JDBC driver I used I am still having the performance problem. If so, many O/R mappings cannot be used because I believe most of them if not all use the precompiled statement.
Best regards
Edmond

Edmond,
Most JDBC drivers for MS SQL (and I think this includes all the drivers you tested) use sp_executesql to execute PreparedStatements. This is a pretty good solution as the driver doesn't have to keep any information about the PreparedStatement locally, the server takes care of all the precompiling and caching. And if the statement isn't already precompiled, this is also taken care of transparently by SQL Server.
The problem with this approach is that all names in the query must be fully qualified. This means that the driver has to parse the query you are submitting and make all names fully qualified (by prepending a db name and schema). This is why creating a PreparedStatement takes so much using these drivers (and why it does so every time you create it, even though it's the same PreparedStatement).
However, the speed advantage of PreparedStatements only becomes visible if you reuse the statement a lot of times.
As about why the PreparedStatement with no placeholder is much faster, I think is because of internal optimisations (maybe the statement is run as a plain statement (?) ).
As a conclusion, if you can reuse the same PreparedStatement, then the performance hit is not so high. Just ignore it. However, if the PreparedStatement is created each time and only used a few times, then you might have a performance issue. In this case I would recommend you try out the jTDS driver ( http://jtds.sourceforge.net ), which uses a completely different approach: temporary stored procedures are created for PreparedStatements. This means that no parsing is done by the driver and PreparedStatement caching is possible (i.e. the next time you are preparing the same statement it will take much less as the previously submitted procedure will be reused).
Alin.

Problem with sql statement after migration

Hi, recently I copied a production database X to a remote location, to serve as a test upgrade database. Original is 32 bit and test database is 64 bit. After migration it turned out that one sql statement consumed much more CPU during EXECUTE phase than on production database. Explain plans are the same. Why CPU is different? How can I trace the problem?
ORIGINAL SERVER TKPROF OUTPUT:
call count cpu elapsed disk query current rows
Parse 1 0.12 0.33 0 0 2 0
Execute 1 542.60 2165.29 2 6 0 0
Fetch 0 0.00 0.00 0 0 0 0
total 2 542.73 2165.62 2 6 2 0
Misses in library cache during parse: 1
Optimizer mode: ALL_ROWS
Parsing user id: 66 (SPB)
Rows     Row Source Operation
      0 SORT UNIQUE (cr=6 pr=2 pw=0 time=719909655 us)
      0   FILTER (cr=6 pr=2 pw=0 time=719909615 us)
      2    NESTED LOOPS (cr=0 pr=0 pw=0 time=719895787 us)
      2     HASH JOIN (cr=0 pr=0 pw=0 time=719881575 us)
      2      NESTED LOOPS (cr=0 pr=0 pw=0 time=1210563833 us)
      2       NESTED LOOPS (cr=0 pr=0 pw=0 time=1210549798 us)
      2        NESTED LOOPS (cr=0 pr=0 pw=0 time=1210519159 us)
330261         REMOTE TWFLPROCESSINSTANCES (cr=0 pr=0 pw=0 time=29176788 us)
      2         REMOTE TAMCONTRACTEDITEMD (cr=0 pr=0 pw=0 time=1177866001 us)
      2        REMOTE TAMPROPVALUES (cr=0 pr=0 pw=0 time=30580 us)
      2       REMOTE TREPVALUELISTITEMS (cr=0 pr=0 pw=0 time=13973 us)
24542745      REMOTE TAMCONTRACTEDITEMD (cr=0 pr=0 pw=0 time=638116004 us)
      2     REMOTE TWFLPROCESSINSTANCES (cr=0 pr=0 pw=0 time=14154 us)
      2    INDEX RANGE SCAN IDX_PROCES_ID (cr=6 pr=2 pw=0 time=13706 us)(object id 546934)
Rows     Execution Plan
      0 INSERT STATEMENT   MODE: ALL_ROWS
      0   SORT (UNIQUE)
      0    FILTER
      2     NESTED LOOPS
      2      HASH JOIN
      2       NESTED LOOPS
      2        NESTED LOOPS
      2         NESTED LOOPS
330261          REMOTE OF 'TWFLPROCESSINSTANCES' (REMOTE)
                     [T6STDBY.WR]
                    SELECT /*+ OPAQUE_TRANSFORM */ "ID","STATUS","STARTDATE"
                    FROM "DIALOG6"."TWFLPROCESSINSTANCES" "PI" WHERE
                      "STATUS"<>'A' AND "STATUS"<>'N' AND "STARTDATE">=:1-30
      2          REMOTE OF 'TAMCONTRACTEDITEMD' (REMOTE)
                     [T6STDBY.WR]
                    SELECT /*+ OPAQUE_TRANSFORM */ "CITEM_ID","ROOTCITEM_ID",
                    "PRODUCT_ID","CONTRACT_ID" FROM
                    "DIALOG6"."TAMCONTRACTEDITEMD" "CI" WHERE "PRODUCT_ID"=
                      934 AND "CONTRACT_ID"=:1
      2         REMOTE OF 'TAMPROPVALUES' (REMOTE) [T6STDBY.WR]
                   SELECT /*+ OPAQUE_TRANSFORM */ "PROPERTY_ID",
                   "VALUEOBJECT_ID","CONTRACTEDITEM_ID","DTO","ISVALID"
                   FROM "DIALOG6"."TAMPROPVALUES" "TV" WHERE "DTO" IS NULL
                   AND "ISVALID"='Y' AND "CONTRACTEDITEM_ID"=:1 AND
                     "PROPERTY_ID"=930326
      2        REMOTE OF 'TREPVALUELISTITEMS' (REMOTE) [T6STDBY.WR]
                  SELECT /*+ OPAQUE_TRANSFORM */ "ID","NAME" FROM
                    "DIALOG6"."TREPVALUELISTITEMS" "TVL" WHERE "ID"=:1
24542745       REMOTE OF 'TAMCONTRACTEDITEMD' (REMOTE) [T6STDBY.WR]
                 SELECT /*+ OPAQUE_TRANSFORM */ "CITEM_ID","CONTRACT_ID" FROM
                   "DIALOG6"."TAMCONTRACTEDITEMD" "CINAD"
      2      REMOTE OF 'TWFLPROCESSINSTANCES' (REMOTE) [T6STDBY.WR]
                SELECT /*+ OPAQUE_TRANSFORM */ "ID" FROM
                "DIALOG6"."TWFLPROCESSINSTANCES" "SYS_ALIAS_1" WHERE "ID"=
                  :1
      2     INDEX   MODE: ANALYZED (RANGE SCAN) OF 'IDX_PROCES_ID'
                (INDEX)
REMOTE SERVER TKPROF OUTPUT:
call count cpu elapsed disk query current rows
Parse 1 0.15 0.15 0 0 2 0
Execute      1   1210.69    3831.70          0          6          0           0
Fetch 0 0.00 0.00 0 0 0 0
total 2 1210.85 3831.85 0 6 2 0
Misses in library cache during parse: 1
Optimizer mode: ALL_ROWS
Parsing user id: 66 (SPB)
Rows     Row Source Operation
      0 SORT UNIQUE (cr=6 pr=0 pw=0 time=681448432 us)
      0   FILTER (cr=6 pr=0 pw=0 time=681448397 us)
      2    NESTED LOOPS (cr=0 pr=0 pw=0 time=681448144 us)
      2     HASH JOIN (cr=0 pr=0 pw=0 time=681443808 us)
      2      NESTED LOOPS (cr=0 pr=0 pw=0 time=168646003 us)
      2       NESTED LOOPS (cr=0 pr=0 pw=0 time=168636029 us)
      2        NESTED LOOPS (cr=0 pr=0 pw=0 time=168580989 us)
327667         REMOTE TWFLPROCESSINSTANCES (cr=0 pr=0 pw=0 time=9177863 us)
      2         REMOTE TAMCONTRACTEDITEMD (cr=0 pr=0 pw=0 time=158011360 us)
      2        REMOTE TAMPROPVALUES (cr=0 pr=0 pw=0 time=55007 us)
      2       REMOTE TREPVALUELISTITEMS (cr=0 pr=0 pw=0 time=9947 us)
24542745      REMOTE TAMCONTRACTEDITEMD (cr=0 pr=0 pw=0 time=3509657414 us)
      2     REMOTE TWFLPROCESSINSTANCES (cr=0 pr=0 pw=0 time=4309 us)
      2    INDEX RANGE SCAN IDX_PROCES_ID (cr=6 pr=0 pw=0 time=160 us)(object id 552075)
Rows     Execution Plan
      0 INSERT STATEMENT   MODE: ALL_ROWS
      0   SORT (UNIQUE)
      0    FILTER
      2     NESTED LOOPS
      2      HASH JOIN
      2       NESTED LOOPS
      2        NESTED LOOPS
      2         NESTED LOOPS
327667          REMOTE OF 'TWFLPROCESSINSTANCES' (REMOTE)
                     [T6STDBY.WR]
                    SELECT /*+ OPAQUE_TRANSFORM */ "ID","STATUS","STARTDATE"
                    FROM "DIALOG6"."TWFLPROCESSINSTANCES" "PI" WHERE
                      "STATUS"<>'A' AND "STATUS"<>'N' AND "STARTDATE">=:1-30
      2          REMOTE OF 'TAMCONTRACTEDITEMD' (REMOTE)
                     [T6STDBY.WR]
                    SELECT /*+ OPAQUE_TRANSFORM */ "CITEM_ID","ROOTCITEM_ID",
                    "PRODUCT_ID","CONTRACT_ID" FROM
                    "DIALOG6"."TAMCONTRACTEDITEMD" "CI" WHERE "PRODUCT_ID"=
                      934 AND "CONTRACT_ID"=:1
      2         REMOTE OF 'TAMPROPVALUES' (REMOTE) [T6STDBY.WR]
                   SELECT /*+ OPAQUE_TRANSFORM */ "PROPERTY_ID",
                   "VALUEOBJECT_ID","CONTRACTEDITEM_ID","DTO","ISVALID"
                 FROM "DIALOG6"."TAMPROPVALUES" "TV" WHERE "DTO" IS NULL
                   AND "ISVALID"='Y' AND "CONTRACTEDITEM_ID"=:1 AND
                     "PROPERTY_ID"=930326
      2        REMOTE OF 'TREPVALUELISTITEMS' (REMOTE) [T6STDBY.WR]
                  SELECT /*+ OPAQUE_TRANSFORM */ "ID","NAME" FROM
                    "DIALOG6"."TREPVALUELISTITEMS" "TVL" WHERE "ID"=:1
24542745       REMOTE OF 'TAMCONTRACTEDITEMD' (REMOTE) [T6STDBY.WR]
                 SELECT /*+ OPAQUE_TRANSFORM */ "CITEM_ID","CONTRACT_ID" FROM
                   "DIALOG6"."TAMCONTRACTEDITEMD" "CINAD"
      2      REMOTE OF 'TWFLPROCESSINSTANCES' (REMOTE) [T6STDBY.WR]
                SELECT /*+ OPAQUE_TRANSFORM */ "ID" FROM
                "DIALOG6"."TWFLPROCESSINSTANCES" "SYS_ALIAS_1" WHERE "ID"=
                  :1
      2     INDEX   MODE: ANALYZED (RANGE SCAN) OF 'IDX_PROCES_ID'
                (INDEX)Edited by: Przemek P on 2012-01-27 07:40
Edited by: Przemek P on 2012-01-27 07:50

Could you please edit your post and use \ tags so the output is more user friendly?
Type: \Your Code Here\It will display as:Your Code Here

Strange problem with SQL-Statement

Hi guys,
i encountered a strange problem trying to execute a statement
like this:
Insert Into Data Set Text = 'Exception-Type: {0} Message: {1}';
I'm using createStatement and executeUpdate.
But the message stored looks like:
Exception-Type: 0{} Message: 1{}
I'm using MySQL 3.23. if I use ODBC everthing is o.k.!?
Any suggestions
Thanks

I found a solution,
i've thought it through once again. the sql-parsing
mechanism seemed to be the right direction.
i guess the odbc mysql implementation does a sql-statement
conversion, so the string is stored correctly into the
database. the jdbc-implementation doesn't.
if i replace { with \{ it works. the backlash can be used
to masquerade some characters in mysql.
my application is database independent (or it should be)
so i wonder if this works with oracle , db2 etc.
mike

Problem with SQL Statement for Result Filtering

Dear Visual Composer Experts,
Here is another Question from me: I have a SQL Query that is working as the data service
Select AB.AgingBandID, AB.AgingBand,
Sum(Case when priority='Emergency' then '1' Else 0 End) as [Emergency],
Sum(Case when priority='Ugent' then '1' Else 0 End) as Ugent,
Sum(Case when priority='High' then '1' Else 0 End) as High,
Sum(Case when priority='Medium' then '1' Else 0 End) as Medium,
Sum(Case when priority='Low' then '1' Else 0 End) as Low
from DimAgingBand AB left outer join
(Select AgingBandID , priority , yeardesc
from vNotifications where YearDesc = (select year(getdate())-1)) as vN
on AB.AgingBandID=vN.AgingBandID
where AB.AgingBandID<>'1'
Group by AB.AgingBandID, AB.AgingBand
Order by AB.AgingBandID
That would return me a table as in the following:
     Agingband     E     U     H     M     L
     < 1week     0     0     0     0     1
     1 - 2 weeks     0     0     0     0     0
     2 - 4weeks     0     0     0     0     1
> 1month     8     2     1     1     6
Now that I would like to add some parameters to filter the result, so I modify the query and put it in the SQL Statement input port of the same data service. The query is like this:
     "Select AB.AgingBandID, AB.AgingBand,Sum(Case when priority='Emergency' then '1' Else 0 End) as [Emergency],Sum(Case when priority='Ugent' then '1' Else 0 End) as Ugent,Sum(Case when priority='High' then '1' Else 0 End) as High,Sum(Case when priority='Medium' then '1' Else 0 End) as Medium,Sum(Case when priority='Low' then '1' Else 0 End) as Low from DimAgingBand AB left outer join (Select AgingBandID , priority , yeardesc from vNotifications where YearDesc like '2009%' and Branch like '" & if(STORE@selectedBranch=='ALL', '%', STORE@selectedBranch) & "' and MainWorkCentre like '%') as vN on AB.AgingBandID=vN.AgingBandID where AB.AgingBandID<>'1' Group by AB.AgingBandID, AB.AgingBand Order by AB.AgingBandID"
However this input port query keeps giving me error as NullPointerException. I have actually specified a condition where the query will run if only STORE@selectedBranch != u2018u2019.
I have other filtering queries working but they are not as complicated query as this one. Could it be possible that query in the input port cannot handle left outer join?
Could it be anything else?
Help is very much appreciated.
Thanks & Regard,
Sarah

Hi,
Thank you very much for your replys. I've tested if the dynamic value of the condition is integer, it's OK
But if the ClassID type is not integer, it's string, I write a SQL Statement like:
"Select DBADMIN.Class.ClassName from DBADMIN.Class where DBADMIN.Class.ClassID = '1' "
or with dynamic condition:
"Select DBADMIN.Class.ClassName from DBADMIN.Class where DBADMIN.Class.ClassID = '"&@ClassID&"'"
or I write the SQL Statement for insert/update/delete data,
I always have errors.
I've tested if the dynamic value of the condition is integer, it's OK
Do you know this problem ?
Thank you very much & kindly regards,
Tweety

Problem constructing SQL statement dynamically

One of our system requirements is to generate a report dynamically based on certain search crieria. I implemented the report by creating a dynamic page. The dynamic page receive the search criteria as page parameters.
Inside the dynamic page, I have an oracle (PL/SQL) section where I built the SQL statement dynamically based on the criteria.
The opening of the cursor, selecting and displaying of the data works fine - no problem there. The only problem I have is with the construction of the where-clause. It almost seems as if the "if-statements" are completely ignored!!
Take for example the following extract that makes use of the page parameter "pSDate" that contains the start date:
     whereClause := 'where claimant.kUserName = Claims.fClaimantUserName';
     if rTrim(ltrim(to_char(:pSDate))) = '' then
     whereClause := whereClause||' and Claims.CaptureDate >= to_date('''||:pSDate||''')';
     end if;
After execution of the above the value of "whereClause" is always "where claimant.kUserName = Claims.fClaimantUserName" ... regardless of whether pSDate is populated or not. In other words, I can't get the "then" part of this statement to execute!
By printing out the value of pSDate I verified that it indeed contains/not contains a value. I experimented with various functions (ltrim, to_char...) without any success. I even changed the operator in the condition to <> and it still does not work. I find this really weird.
This is probably something really stupid but being new to PL/SQL I am missing it. Is there anybody out there that knows how I can solve this?

Tried it and unfortunatly it does not work.
By specifying the ":" you inform the compiler that it's a parameter to the "dynamic page". As a result it's added as a bind variable of the page. That is visible on the "Customization Form Display Options"-tab.
The dynamic page is "called" from a stored procedure which "pass" values to these parameters. The passing seems to work fine since, when I "print" the parameter value in the page e.g.
htp.p('pSDate: '||:pSDate)
the correct value is displayed.
The question remains, why does it not work when I use this parameter/bind variable in a condition that forms part of a if-statement?
I have tried to assign the parameter to a tmp variable of type varchar2(100) just before the if-statement. Using the tmp variable in the if-statement makes no difference.
Pleeeease can someone help me

Problems with SQL statement ..Urgent Help required

Hi
I have a table with something like this
::: inidicates column seperation
col1 ::: col2 ::: col3 ::: col4
ab,abc,abvfg,fghy ::: x1 ::: y1 ::: z1
ba,mkl,huji,kill ::: x2 ::: y2 ::: z2
I want to get something like this :
col1 ::: col2 ::: col3 ::: col4
ab ::: x1 ::: y1 ::: z1
abc ::: x1 ::: y1 ::: z1
abvfg ::: x1 ::: y1 ::: z1
fghy ::: x1 ::: y1 ::: z1
and so on......
Either a procedure or a single sql statement is fine..
PS: the first column can have any number of words separated with comma (,)
Please advise
Thanks in advance

create table kris (
a varchar2(200),
b varchar2(20),
c varchar2(20),
d varchar2(20));
insert into kris values('ab,abc,abvfg,fghy', 'x1','y1','z1');
insert into kris values('ba,mkl,huji,kill', 'x2','y2','z2');
select substr(','||a||',', instr(','||a||',', ',', 1, rn) +1,
     instr(','||a||',', ',', 1, rn+1) - instr(','||a||',', ',', 1, rn) -1)
     v,b,c,d
from (select length(a) - length(translate(a, '1,', '1')) +1 L,
     a, b, c, d from kris) t, (select rownum rn from cat) c
     where c.rn <= t.L
order by a, rn;
V B C D
ab x1 y1 z1
abc x1 y1 z1
abvfg x1 y1 z1
fghy x1 y1 z1
ba x2 y2 z2
mkl x2 y2 z2
huji x2 y2 z2
kill x2 y2 z2

Performance Issie in SQL Statement Using Hints

Hi,
I am using Oracle version 10G.
I am using a INSERT statement as:
INSERT /*+ PARALLEL (TEST,4) */ INTO TEST(COL1,COL2,COL3) SELECT COL1,COL2,COL3 FROM DUMMY;
For increasing the Performance I am using the above statement.
Will the usage of Hints increase the Performance?
Any help will be highly needful.
Thanks and Regards

user598986 wrote:
I am using Oracle version 10G.
I am using a INSERT statement as:
INSERT /*+ PARALLEL (TEST,4) */ INTO TEST(COL1,COL2,COL3) SELECT COL1,COL2,COL3 FROM DUMMY;
For increasing the Performance I am using the above statement.
Will the usage of Hints increase the Performance?The way you're asking the question suggests that you're not sure what this particular hint is supposed to imply.
Using this hint suggests that you want to take advantage of direct-path parallel DML operations. Note that you explicitly need to enable parallel DML in your session, it is disabled by default, because it has some significant implications and restrictions. You should think about parallelizing the query on DUMMY, too, if it is not marked as PARALLEL in the dictionary, because otherwise you're combining a parallel DML operation with a serial query which might not be that efficient.
Note that there is no general answer to the question if this particular hint will actually increase the performance of the DML statement. There are many things to consider, among them are if your system scales reasonably with parallel operations and if the underlying object structure actually allows to benefit from the parallel operation. There are cases where a serial operation might be faster than a parallel operation.
For more information about direct-path and parallel execution, its implications and restrictions, see the documentation:
http://download.oracle.com/docs/cd/B28359_01/server.111/b28286/statements_9014.htm#i2163698
http://download.oracle.com/docs/cd/B28359_01/server.111/b28313/usingpe.htm#CACEJACE
You can use the EXPLAIN PLAN and the DBMS_XPLAN.DISPLAY function to get the execution plan of your statement that shows you what kind of parallel operations the optimizer estimates to perform.
Regards,
Randolf
Oracle related stuff blog:
http://oracle-randolf.blogspot.com/
SQLTools++ for Oracle (Open source Oracle GUI for Windows):
http://www.sqltools-plusplus.org:7676/
http://sourceforge.net/projects/sqlt-pp/

Problems storing sql statement in database

Hi,
I'm writing an application which needs to select a sql statement from an oracle database. The statement should look something like this:
"select id,name,... .from table 1 where id = " + ID + " and name = " + NAME
however when I select this from the database it give a sql error - invalid sql statement, I've tried putting it in the database without quotes, with single quotes etc but I can't seem to get it working.
I'd much appreciate it if any one had any ideas on this.
thanks.

debug it: output the sql string to a println or
JTextArea. See what is in it (and test it) . Must be a
null value in the variables.The variables are set previously in the code
I outputted the sql string to the screen but the variable names are not being replaced by the values.

Problem with sql statement

hi
i have a strange problem....when i run the following code with the 'if' statement commented out then it works fine. but when i have to search a particular name, say roy , then i include that if statement. but now it gives an error like no such data found.
if the return typ of result.getString(2) is a string type then i wonder what is the problem. i enclose the code below.
import java.sql.*;
import java.io.*;
public class checkDoc
     public static void main(String args[])
          String name;
          ResultSet result;
          try
               name= "roy";
               Class.forName("sun.jdbc.odbc.JdbcOdbcDriver");
               Connection con;
               con = DriverManager.getConnection("jdbc:odbc:MyDataSource","","");
               Statement stat = con.createStatement();
               result = stat.executeQuery("Select * from Details1");
               while (result.next())
/* problem here*/
/*     if(result.getString(2).equals(name))
     */               System.out.println(result.getString(2));
          catch(Exception e)
               System.out.println("Error"+e);

thanks alot for that help...but now there is another silly problem!..
i following is perhaps the simplest code possible for insering data into an ACCESS database.(simplest possible with just one column by the name Doc_name and table name is docdetails)...but it gives the error...
"java.sql.SQLException:syntax error or in INSERT INTO statement"
someone plz help!
import java.sql.*;
public class checkDoc4
     public static void main(String args[])
          ResultSet rs;
          try
               Class.forName("sun.jdbc.odbc.JdbcOdbcDriver");
               Connection con;
               con = DriverManager.getConnection("jdbc:odbc:MyDataSource2","","");
Statement stat = con.createStatement();
                    stat.executeUpdate("INSERT INTO docdetails" +
     "VALUES ('ankur')");
          catch(Exception e)
               System.out.println("Error"+e);

Performance problem with MERGE statement

Version : 11.1.0.7.0
I have an insert statement like following which is taking less than 2 secs to complete and inserts around 4000 rows:
INSERT INTO sch.tab1
          (c1,c2,c3)
SELECT c1,c2,c3
   FROM sch1.tab1@dblink
WHERE c1 IN (SELECT c1 FROM sch1.tab2@dblink);I wanted to change it to a MERGE statement just to avoid duplicate data. I changed it to following :
MERGE INTO sch.tab1 t1
USING (SELECT c1,c2,c3
   FROM sch1.tab1@dblink
WHERE c1 IN (SELECT c1 FROM sch1.tab2@dblink) t2
ON (t1.c1 = t2.c1)
WHEN NOT MATCHED THEN
INSERT (t1.c1,t1.c2,t1.c3)
VALUES (t2.c1,t2.c2,t2.c3);The MERGE statement is taking more than 2 mins (and I stopped the execution after that). I removed the WHERE clause subquery inside the subquery of the USING section and it executed in 1 sec.
If I execute the same select statement with the WHERE clause outside the MERGE statement, it takes just 1 sec to return the data.
Is there any known issue with MERGE statement while implementing using above scenario?

riedelme wrote:
Are your join columns indexed?
Yes, the join columns are indexed.
You are doing a remote query inside the merge; remote queries can slow things down. Do you have to select all thr rows from the remote table? What if you copied them locally using a materialized view?Yes, I agree that remote queries will slow things down. But the same is not happening while select, insert and pl/sql. It happens only when we are using MERGE. I have to test what happens if we use a subquery refering to a local table or materialized view. Even if it works, I think there is still a problem with MERGE in case of remote subqueries (atleast till I test local queries). I wish some one can test similar scenarios so that we can know whether it is a genuine problem or some specific problem from my side.
>
BTW, I haven't had great luck with MERGE either :(. Last time I tried to use it I found it faster to use a loop with insert/update logic.
Edited by: riedelme on Jul 28, 2009 12:12 PM:) I used the same to overcome this situation. I think MERGE needs to be still improved functionally from Oracle side. I personally feel that it is one of the robust features to grace SQL or PL/SQL.

Again the SUM and DISTINCT Problem (With SQL Statement)

Here is the actual SQL,
select
NVL(l.pr_cust_prod,'zzzzz') A,
NVL(l.pr_prod_code,'zzzzz') B,
NVL(l.pr_lot_no,'zzzzz') C,
NVL(l.db_waf_qty,'99999') lotQty,
NVL(t.db_cust_po,'zzzzz') poNum,
NVL(p.oe_po_line,'99999') poLineNum,
i.db_shpmt_no invNum,
from
db_invoice i, db_line_lot l, db_inv_item t, db_prod_line p
where
l.db_shpmt_no like 'U388%'
and i.db_invo_date > sysdate - 500
and l.pr_run_type = 'PR'
and l.db_shpmt_no = i.db_shpmt_no
and l.db_shpmt_no = t.db_shpmt_no
and l.db_item_no = t.db_item_no
and p.db_ref_numa1 = t.db_tsmc_po
and l.db_shpmt_no not in
(select distinct db_shpmt_no from ftp_cust_ship
where DB_MAST_ACCT='U388' and sys_name='SENT')
and l.pr_prod_code like '%Z_'
group by l.pr_lot_no, l.db_line_numb, l.pr_cust_prod, l.pr_prod_code,
t.db_cust_po, i.db_delv_cod2, l.db_lot_desc, i.db_shpmt_no,
p.oe_po_line
order by lotNum
the problem is that if I dont use SELECT DISTINCT at beginning
the result is like
F , LOTQTY , PONUM , POLINENUM
DP8568.00,25 ,BVI10080 ,1
DP8568.00,25 ,BVI10080 ,2
DP8568.00,25 ,BVI10080 ,1
DP8568.00,25 ,BVI10080 ,1
DP8568.00,25 ,BVI10080 ,2
DP8568.00,25 ,BVI10080 ,1
DP8568.00,25 ,BVI10080 ,1
DP8568.00,25 ,BVI10080 ,2 //pleae note the QTY and POLINENUM
Then if Add SELECT DISTINCT, the result will be
F ,LOTQTY , PONUM , POLINENUM
DP8568.00,25 ,BVI10080,1
DP8568.00,25 ,BVI10080,2
but I change the NVL(l.db_waf_qty,'99999') lotQty, statement to NVL((SUM( l.db_waf_qty)),'99999') lotQty,
this will give me the result
F ,LOTQTY, PONUM , POLINENUM
DP8568.00,75 , BVI10080 ,2
DP8568.00,125 , BVI10080 ,1
Which I excepted the result as
F ,LOTQTY, PONUM , POLINENUM
DP8568.00,25 , BVI10080 ,2
DP8568.00,25 , BVI10080 ,1
And Advise??
Thanks a lot
Jon

based on the data set you given
SQL>
SQL> With t As
2 (
3 Select 'DP8568' c1 ,0.25 c2,'BVI10080' c3,1 c4 From dual Union All
4 Select 'DP8568' ,0.25 ,'BVI10080' ,2           From dual Union All
5 Select 'DP8568' ,0.25 ,'BVI10080' ,1           From dual Union All
6 Select 'DP8568' ,0.25 ,'BVI10080' ,1           From dual Union All
7 Select 'DP8568' ,0.25 ,'BVI10080' ,2           From dual Union All
8 Select 'DP8568' ,0.25 ,'BVI10080' ,1           From dual Union All
9 Select 'DP8568' ,0.25 ,'BVI10080' ,1           From dual Union All
10 Select 'DP8568' ,0.25 ,'BVI10080' ,2           From dual
11 )
12 Select c1,
13         sum(c2) lot_qty,
14         c3,
15         c4
16 From
17 (
18    Select c1, --Sum(c2) over (Partition By c1,c4 Order By c4)
19           row_number() over (Partition By c1,c4 Order By c4) rn,
20           c2,
21           c3,
22           c4
23    From t
24 ) Where rn = 1
25 Group By c1, c3, c4;
C1        LOT_QTY C3               C4
DP8568       0.25 BVI10080          2
DP8568       0.25 BVI10080          1
SQL>

Problems with SQL statement

Here is the problem that I have .. I have a table that gets updated every 30 minutes.. and the columns are:
Day Shift MDate
The data is as follows:
wed M 09012003
wed N 09012003
thu M 09022003
thu N 09022003
I want a SQL or PL/SQL that can help me in grabbing the distinct MDate where I have the instance of both M and N in the shift.
For the above data, I want to get distinct MDates as 09012003
09022003
But here is the problem; at some point in time say in the evening.. the data looks like this
wed M 09012003
wed N 09012003
thu M 09022003
For the above data, I want to get distinct MDates as 09012003
This is because "day" column with "thu" does not have both the instances of M and N in the "shift" column
Any help in this regard is highly appreciated..
Thanks in advance

Use 'group by' and 'having':
create table ttest(Day varchar2(3), Shift varchar2(1), MDate varchar(8));
insert into ttest values('wed', 'M', '09012003');
insert into ttest values('wed', 'N', '09012003');
insert into ttest values('thu', 'M', '09022003');
insert into ttest values('thu', 'N', '09022003');
select distinct mdate from ttest group by mdate having count(1) > 1;
MDATE
09012003
09022003
delete from ttest;
insert into ttest values('wed', 'M', '09012003');
insert into ttest values('wed', 'N', '09012003');
insert into ttest values('thu', 'M', '09022003');
select distinct mdate from ttest group by mdate having count(1) > 1;
MDATE
09012003

Performance Problem in SQL Statement

Similar Messages

Maybe you are looking for