Top Ten SQL Performance Trouble Clues
DB:
Oracle Database 11g Enterprise Edition Release 11.1.0.7.0 - 64bit Production
PL/SQL Release 11.1.0.7.0 - Production
TNS for Linux: Version 11.1.0.7.0 - Production
NLSRTL Version 11.1.0.7.0 - Production
OS: Red Hat Enterprise Linux 5
At the risk of being over-simplistic, what would be the Top 5 or 10 things one looks for in an explain plan or execution plan that suggests 'performance trouble'? ...
user601798 wrote:
Well, one has to start somewhere and whether its looking at the execution plan or trace file, there must be some rules of thumb... Indeed, we all look 'somewhere' first or notice something in the plan or trace that seems to point to the problematic issue.
And granted, each tuning issue is unique but, based on experience, I can say that quite a few of my tuning issues have been resolved by simply using NESTED_LOOPS and having the first driving table eliminate as many rows as possible from the final result set.
And based on that, I'd say you tend to work on OLTP systems.
If you were a data warehouse DBA, you might have said:
"And granted, each tuning issue is unique but, based on experience, I can say that quite a few of my tuning issues have been resolved by simply making sure I have bitmap indexes defined on columns that appear in predicates, and making sure sure I have allocated enough memory for efficient HASH joins to take place."
So, it depends on your perspective.
That's why performance optimization methods, such as Method-R, focus on identifying the most important, business critical functions, profiling them, understanding where the time is being spent, and focusing the tuning efforts there. In this way, there are no rules of thumb, and there is no guesswork. If you have a performance problem, Method-R will point you to the source of the problem. It's then up to you to understand enough about Oracle to know how to proceed.
-Mark
Similar Messages
-
Urgt:How to identify TOP 5 SQL using more CPU time without using statspack
How to identify the TOP 5 SQL queries which are consuming more CPU time during the timespan of 24 hours for entire database. There are N number of users who have issued sql queries, out of which few users have disconnected and few user are still connected to the database(I need to consider all sessions). My database version is 9.0.1. I don't want to use statspack. Is there any way to identify TOP 5 SQL consuming more CPU time without affecting the performance of the database. Can any one help me on this? Its urgent!!. I don't want to use statspack becasue it degrades the performance and i cannot afford to run it for 24 hours.
You don't run statspack for 24 hours, that would infact be pointless...
Run statspack for say a 10 or 15 minute window during a particular "busy" period on your database. If you run it for longer then all the useful data will have been "averaged out". You want a snapshot in time, not an average over the day. -
How to identify TOP 5 SQL consuming more CPU time without using statspack
How to identify the TOP 5 SQL queries which are consuming more CPU time during the timespan of 24 hours for entire database. There are N number of users who have issued sql queries, out of which few users have disconnected and few user are still connected to the database(I need to consider all sessions). My database version is 9.0.1. I don't want to use statspack. Is there any way to identify TOP 5 SQL consuming more CPU time without affecting the performance of the database. Can any one help me on this? Its urgent!!.
My database version is 9.0.1. I don't want
to use statspack. Is there any way to identify TOP 5
SQL consuming more CPU time without affecting the
performance of the database. Can any one help me on
this? Its urgent!!.I cant understand why you dont want to use statspack. As it is very easy to use and very quick to identify bottlenecks. Try using statspack. Its really wonderful.
Particularly in your case statspack is the best.
Regards,
Sanju. -
How can I improve below SQL performance.
Hi,
How can I improve below SQL performance. This SQL consumes CPU and occures wait events. It is running every 10 seconds. When I look at the session information from Enterprise Manager I can see that "Histogram for Wait Event: PX Deq Credit: send blkd"
I created some indexes. I heard that the indexes are not used when there is a NULL but when I checked the xecution plan It uses index.
SELECT i.ID
FROM EXPRESS.invoices i
WHERE i.nbr IS NOT NULL
AND i.EXTRACT_BATCH IS NULL
AND i.SUB_TYPE='COD'
Explain Plan from Toad
SELECT STATEMENT CHOOSECost: 77 Bytes: 6,98 Cardinality: 349
4 PX COORDINATOR
3 PX SEND QC (RANDOM) SYS.:TQ10000 Cost: 77 Bytes: 6,98 Cardinality: 349
2 PX BLOCK ITERATOR Cost: 77 Bytes: 6,98 Cardinality: 349
1 INDEX FAST FULL SCAN INDEX EXPRESS.INVC_TRANS_INDX Cost: 77 Bytes: 6,98 Cardinality: 349
Execution Plan from Sqlplus
| Id | Operation | Name | Rows | Bytes | Cost | TQ |IN-OUT| PQ Distrib |
| 0 | SELECT STATEMENT | | 349 | 6980 | 77 | | | |
| 1 | PX COORDINATOR | | | | | | | |
| 2 | PX SEND QC (RANDOM) | :TQ10000 | 349 | 6980 | 77 | Q1,00 | P->S | QC (RAND) |
| 3 | PX BLOCK ITERATOR | | 349 | 6980 | 77 | Q1,00 | PCWC | |
|* 4 | INDEX FAST FULL SCAN| INVC_TRANS_INDX | 349 | 6980 | 77 | Q1,00 | PCWP | |
Predicate Information (identified by operation id):
4 - filter("I"."NBR" IS NOT NULL AND "I"."EXTRACT_BATCH" IS NULL AND "I"."SUB_TYPE"='COD')
Note
- 'PLAN_TABLE' is old version
- cpu costing is off (consider enabling it)
Statistics
141 recursive calls
0 db block gets
5568 consistent gets
0 physical reads
0 redo size
319 bytes sent via SQL*Net to client
458 bytes received via SQL*Net from client
1 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
0 rows processed
Instance Efficiency Percentages (Target 100%)
Buffer Nowait %: 100.00
Redo NoWait %: 100.00
Buffer Hit %: 99.70
In-memory Sort %: 100.00
Library Hit %: 99.81
Soft Parse %: 99.77
Execute to Parse %: 63.56
Latch Hit %: 90.07
Parse CPU to Parse Elapsd %: 0.81
% Non-Parse CPU: 98.88
Top 5 Timed Events
Event Waits Time(s) Avg Wait(ms) % Total Call Time Wait Class
latch: library cache 12,626 16,757 1,327 62.6 Concurrency
CPU time 5,712 21.Mar
latch: session allocation 1,848,987 1,99 1 07.Nis Other
PX Deq Credit: send blkd 1,242,265 981 1 03.Tem Other
PX qref latch 1,405,819 726 1 02.Tem Other
The database version is 10.2.0.1 but we haven't installed the patch 10.2.0.5. yet.
I am waiting your comments.
Thanks in advanceWelcome to the forum.
I created some indexes. I heard that the indexes are not used when there is a NULL but when I checked the xecution plan It uses index. What columns are indexed?
And what do:
select i.sub_type
, count(*)
from express.invoices i
where i.nbr is not null
and i.extract_batch is null
group by i.sub_type; and
select i.sub_type
, count(*)
from express.invoices i
group by i.sub_type; return?
Also, try use the {noformat}{noformat} tag when posting examples/execution plans etc.
See: HOW TO: Post a SQL statement tuning request - template posting for more tuning instructions.
It'll make a big difference:
SELECT i.ID
FROM EXPRESS.invoices i
WHERE i.nbr IS NOT NULL
AND i.EXTRACT_BATCH IS NULL
AND i.SUB_TYPE='COD'
Explain Plan from Toad
SELECT STATEMENT CHOOSECost: 77 Bytes: 6,98 Cardinality: 349
4 PX COORDINATOR
3 PX SEND QC (RANDOM) SYS.:TQ10000 Cost: 77 Bytes: 6,98 Cardinality: 349
2 PX BLOCK ITERATOR Cost: 77 Bytes: 6,98 Cardinality: 349
1 INDEX FAST FULL SCAN INDEX EXPRESS.INVC_TRANS_INDX Cost: 77 Bytes: 6,98 Cardinality: 349
Execution Plan from Sqlplus
| Id | Operation | Name | Rows | Bytes | Cost | TQ |IN-OUT| PQ Distrib |
| 0 | SELECT STATEMENT | | 349 | 6980 | 77 | | | |
| 1 | PX COORDINATOR | | | | | | | |
| 2 | PX SEND QC (RANDOM) | :TQ10000 | 349 | 6980 | 77 | Q1,00 | P->S | QC (RAND) |
| 3 | PX BLOCK ITERATOR | | 349 | 6980 | 77 | Q1,00 | PCWC | |
|* 4 | INDEX FAST FULL SCAN| INVC_TRANS_INDX | 349 | 6980 | 77 | Q1,00 | PCWP | |
Predicate Information (identified by operation id):
4 - filter("I"."NBR" IS NOT NULL AND "I"."EXTRACT_BATCH" IS NULL AND "I"."SUB_TYPE"='COD')
Note
- 'PLAN_TABLE' is old version
- cpu costing is off (consider enabling it)
Statistics
141 recursive calls
0 db block gets
5568 consistent gets
0 physical reads
0 redo size
319 bytes sent via SQL*Net to client
458 bytes received via SQL*Net from client
1 SQL*Net roundtrips to/from client
1 sorts (memory)
0 sorts (disk)
0 rows processed
Instance Efficiency Percentages (Target 100%)
Buffer Nowait %: 100.00
Redo NoWait %: 100.00
Buffer Hit %: 99.70
In-memory Sort %: 100.00
Library Hit %: 99.81
Soft Parse %: 99.77
Execute to Parse %: 63.56
Latch Hit %: 90.07
Parse CPU to Parse Elapsd %: 0.81
% Non-Parse CPU: 98.88
Top 5 Timed Events
Event Waits Time(s) Avg Wait(ms) % Total Call Time Wait Class
latch: library cache 12,626 16,757 1,327 62.6 Concurrency
CPU time 5,712 21.Mar
latch: session allocation 1,848,987 1,99 1 07.Nis Other
PX Deq Credit: send blkd 1,242,265 981 1 03.Tem Other
PX qref latch 1,405,819 726 1 02.Tem Other -
The Top Ten things that BB needs to fix now
BB Fans after having my BB Classic for nearly a month, I am disappointed that there are several issues that need to be resolved immediately with OS 10. I have started a Top Ten list of issues and features that need to be addressed immediately. Please chime in as many of you have the same issues or know of some that I haven't discovered yet. Now here are my Top Ten:
1. Unable to a complete Synch with BB Link, according to BB tech support the issue comes from pictures that are attached to a contact or a Linked In account. We didn't have this issue with BB OS7.
2. Unable to attach a contact to a text message, error shows contact acnnot be shared.
3. Unable to attach a contact to an e-mail, error shows contact cannot be shared.
4. BB Protect doesn't back up and restore as it did in OS7.
5. BB Protect doesn't locate, display a message or play a sound when you are roaming on another network as it did in OS7. It will only locate on your home network, but it won't perform the other features as it did in OS7.
6. When you search for an e-mail, it takes over 10 seconds and it replys "no results". Go to web based e-mail account and perform the same search and the e-mail is found in seconds. OS7 did this well.
7. BB tech support needs to get better, but part of the problem is that they don't have solutions for our concerns.
8. You cannot forward part of a conversion, it will only forward the complete conversation.
9. You cannot bcc yourself to enable you to keep a copy of an e-mail.
10. an e-mail icon needs to be seperate from hub as text messages are, because it takes to long to get to e-mails when accessing them from the hub.
Please everyone chime in as we need to develop a list of our Top Ten issues and features that we desire in our BB10 devices. BB should listen to us as many of us are loyal to BB nd want to see BB survive, improve and grow.Hi, rwb007,
A few thoughts ...
8. You cannot forward part of a conversion, it will only forward the complete conversation.
True, however copy and paste works: http://docs.blackberry.com/en/smartphone_users/deliverables/67769/mba1343759181682.jsp
9. You cannot bcc yourself to enable you to keep a copy of an e-mail.
How to do this:
a. Compose a new e-mail.
b. Tap the menu icon (three dots) at the lowest right corner
c Tap Add Bcc:
10. an e-mail icon needs to be seperate from hub as text messages are, because it takes to long to get to e-mails when accessing them from the hub.
Remember, you can always select the e-mail account on the Hub, so only that e-mail account's messages are displayed. That said, I'm with you on this one - would be nice to have the option, at least - to add an icon to the main screen for separate e-mail accounts as we had on the 9900/OS 7, although I can hear folks reminding me that also defeats the purpose of the Hub. The simplified Hub design leaves me tapping a lot on the glass to separate which account's e-mails I wish to review as sometimes I find it confusing that all e-mails are mixed together in one long list.
Kind regards,
OBBG -
Hi Friends,
I wanted to find the response time for the top 10 transactions in sto3n, I went to the Transaction profile and into the standard option
There it shows me a list of transactions, should I look at the columns total response time (s) or the average reponse time/dialog step (ms)
Can someone guide me on this...
Thanks
AliilaHi Aliila,
If want to see top ten transaction in a particular time interval then go to ST03 ,on the left hand side below worload there will be detailed analysis click it to open the list inside it then choose last minute load and specify the time interval after that click the green tick mark to get the result . On the left hand side you will see Analysis Views below that there will be Transaction Profile tab just click it to open the elements inside it and finally double click the standard. You will see the list of all the transaction runned in that interval then based on ur required for example to get the top 10 transaction on dialog response time you can sort that coloumn in desending order to get the desired result or to get the top 10 transaction based on total response time sort the total response time coloumn in desending order..
To check the performance check Average Response time (ms)
Dialog response time for a particular transaction should be less than 1200(ms) if it is crossing 2000(ms) then check where it is consuming the max time it may be CPU time DB response time etc.if it is a Z report than you have to tune that report.
Regards,
Prashant
Edited by: Prashant Shukla on Oct 13, 2008 10:45 AM
Edited by: Prashant Shukla on Oct 13, 2008 10:46 AM -
How to find out the top ten Change log tables in BW
I want to know the top ten change log tables in terms of size. Do we have a SAP standard table where we have the following fields : "changlog table name", "number of records" field or "data in size" field.
Regards,
Prashant M JHi,
Click on change log table in the ODS/DSO at the top you see the name as /bic* which is the name of the database table which could also be seen at SE11.
If you wana see the requests in this change log use rstsodsrequest
and if you wana see the size in terms of kb, mb DB02 or st04
use the table name /bic* and click on tables and indexes option in history tab
DB02 tells you the sizes of the tables present on database. Incase you are not able to your basis team would help you in that
Thanks and regards
Kiran -
Help needed in SQL performance - Using CASE in SQL statement versus 2 query
Hi,
I have a requirement to find count from a bunch of tables.
The SQL I have gives the count of all members.
I have created 2 queries to find count of active and inactive members.
The key difference is only the active dates.
Each query takes 20 seconds to execute.
I modified the SQL to use CASE statement in the SELECT.
So after the data is fetched the CASE statement will evaluate the active date and gives 2 counts (active and inactive)
Is it advisable to use this approach. Will CASE improve SQL performance ? I have to justify this.
Please let me know your thoughts.
Thanks,
JHi,
If it can be done in single SQL do it in single SQL.
You said:
Will CASE improve SQL performance There can be both cases to prove if the performance is better or worse.
In your case you should tell us how it is.
Regards,
Bhushan -
SQL Performance issue: Using user defined function with group by
Hi Everyone,
im new here and I really could need some help on a weird performance issue. I hope this is the right topic for SQL performance issues.
Well ok, i create a function for converting a date from timezone GMT to a specified timzeone.
CREATE OR REPLACE FUNCTION I3S_REP_1.fnc_user_rep_date_to_local (date_in IN date, tz_name_in IN VARCHAR2) RETURN date
IS
tz_name VARCHAR2(100);
date_out date;
BEGIN
SELECT
to_date(to_char(cast(from_tz(cast( date_in AS TIMESTAMP),'GMT')AT
TIME ZONE (tz_name_in) AS DATE),'dd-mm-yyyy hh24:mi:ss'),'dd-mm-yyyy hh24:mi:ss')
INTO date_out
FROM dual;
RETURN date_out;
END fnc_user_rep_date_to_local;The following statement is just an example, the real statement is much more complex. So I select some date values from a table and aggregate a little.
select
stp_end_stamp,
count(*) noi
from step
where
stp_end_stamp
BETWEEN
to_date('23-05-2009 00:00:00','dd-mm-yyyy hh24:mi:ss')
AND
to_date('23-07-2009 00:00:00','dd-mm-yyyy hh24:mi:ss')
group by
stp_end_stampThis statement selects ~70000 rows and needs ~ 70ms
If i use the function it selects the same number of rows ;-) and takes ~ 4 sec ...
select
fnc_user_rep_date_to_local(stp_end_stamp,'Europe/Berlin'),
count(*) noi
from step
where
stp_end_stamp
BETWEEN
to_date('23-05-2009 00:00:00','dd-mm-yyyy hh24:mi:ss')
AND
to_date('23-07-2009 00:00:00','dd-mm-yyyy hh24:mi:ss')
group by
fnc_user_rep_date_to_local(stp_end_stamp,'Europe/Berlin')I understand that the DB has to execute the function for each row.
But if I execute the following statement, it takes only ~90ms ...
select
fnc_user_rep_date_to_gmt(stp_end_stamp,'Europe/Berlin','ny21654'),
noi
from
select
stp_end_stamp,
count(*) noi
from step
where
stp_end_stamp
BETWEEN
to_date('23-05-2009 00:00:00','dd-mm-yyyy hh24:mi:ss')
AND
to_date('23-07-2009 00:00:00','dd-mm-yyyy hh24:mi:ss')
group by
stp_end_stamp
)The execution plan for all three statements is EXACTLY the same!!!
Usually i would say, that I use the third statement and the world is in order. BUT I'm working on a BI project with a tool called Business Objects and it generates SQL, so my hands are bound and I can't make this tool to generate the SQL as a subselect.
My questions are:
Why is the second statement sooo much slower than the third?
and
Howcan I force the optimizer to do whatever he is doing to make the third statement so fast?
I would really appreciate some help on this really weird issue.
Thanks in advance,
AndiHi,
The execution plan for all three statements is EXACTLY the same!!!Not exactly. Plans are the same - true. They uses slightly different approach to call function. See:
drop table t cascade constraints purge;
create table t as select mod(rownum,10) id, cast('x' as char(500)) pad from dual connect by level <= 10000;
exec dbms_stats.gather_table_stats(user, 't');
create or replace function test_fnc(p_int number) return number is
begin
return trunc(p_int);
end;
explain plan for select id from t group by id;
select * from table(dbms_xplan.display(null,null,'advanced'));
explain plan for select test_fnc(id) from t group by test_fnc(id);
select * from table(dbms_xplan.display(null,null,'advanced'));
explain plan for select test_fnc(id) from (select id from t group by id);
select * from table(dbms_xplan.display(null,null,'advanced'));Output:
PLAN_TABLE_OUTPUT
Plan hash value: 47235625
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
| 0 | SELECT STATEMENT | | 10 | 30 | 162 (3)| 00:00:02 |
| 1 | HASH GROUP BY | | 10 | 30 | 162 (3)| 00:00:02 |
| 2 | TABLE ACCESS FULL| T | 10000 | 30000 | 159 (1)| 00:00:02 |
Query Block Name / Object Alias (identified by operation id):
1 - SEL$1
2 - SEL$1 / T@SEL$1
Outline Data
/*+
BEGIN_OUTLINE_DATA
FULL(@"SEL$1" "T"@"SEL$1")
OUTLINE_LEAF(@"SEL$1")
ALL_ROWS
OPTIMIZER_FEATURES_ENABLE('10.2.0.4')
IGNORE_OPTIM_EMBEDDED_HINTS
END_OUTLINE_DATA
Column Projection Information (identified by operation id):
1 - (#keys=1) "ID"[NUMBER,22]
2 - "ID"[NUMBER,22]
34 rows selected.
SQL>
Explained.
SQL>
PLAN_TABLE_OUTPUT
Plan hash value: 47235625
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
| 0 | SELECT STATEMENT | | 10 | 30 | 162 (3)| 00:00:02 |
| 1 | HASH GROUP BY | | 10 | 30 | 162 (3)| 00:00:02 |
| 2 | TABLE ACCESS FULL| T | 10000 | 30000 | 159 (1)| 00:00:02 |
Query Block Name / Object Alias (identified by operation id):
1 - SEL$1
2 - SEL$1 / T@SEL$1
Outline Data
/*+
BEGIN_OUTLINE_DATA
FULL(@"SEL$1" "T"@"SEL$1")
OUTLINE_LEAF(@"SEL$1")
ALL_ROWS
OPTIMIZER_FEATURES_ENABLE('10.2.0.4')
IGNORE_OPTIM_EMBEDDED_HINTS
END_OUTLINE_DATA
Column Projection Information (identified by operation id):
1 - (#keys=1) "TEST_FNC"("ID")[22]
2 - "ID"[NUMBER,22]
34 rows selected.
SQL>
Explained.
SQL> select * from table(dbms_xplan.display(null,null,'advanced'));
PLAN_TABLE_OUTPUT
Plan hash value: 47235625
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time |
| 0 | SELECT STATEMENT | | 10 | 30 | 162 (3)| 00:00:02 |
| 1 | HASH GROUP BY | | 10 | 30 | 162 (3)| 00:00:02 |
| 2 | TABLE ACCESS FULL| T | 10000 | 30000 | 159 (1)| 00:00:02 |
Query Block Name / Object Alias (identified by operation id):
1 - SEL$F5BB74E1
2 - SEL$F5BB74E1 / T@SEL$2
Outline Data
/*+
BEGIN_OUTLINE_DATA
FULL(@"SEL$F5BB74E1" "T"@"SEL$2")
OUTLINE(@"SEL$2")
OUTLINE(@"SEL$1")
MERGE(@"SEL$2")
OUTLINE_LEAF(@"SEL$F5BB74E1")
ALL_ROWS
OPTIMIZER_FEATURES_ENABLE('10.2.0.4')
IGNORE_OPTIM_EMBEDDED_HINTS
END_OUTLINE_DATA
Column Projection Information (identified by operation id):
1 - (#keys=1) "ID"[NUMBER,22]
2 - "ID"[NUMBER,22]
37 rows selected. -
[sql performance] inline view , group by , max, join
Hi. everyone.
I have a question with regard to "group by" inline view ,
max value, join, and sql performance.
I will give you simple table definitions in order for you
to understand my intention.
Table A (parent)
C1
C2
C3
Table B (child)
C1
C2
C3
C4 number type(sequence number)
1. c1, c2, c3 are the key columns of tabla A.
2. c1, c2, c3, c4 are the key columns of table B.
3. table A is the parent table of Table B.
4. c4 column of table b is the serial number.
(c4 increases from 1 by "1" regarding every (c1,c2,c3)
the following is the simple example of the sql query.
select .................................
from table_a,
(select c1, c2, c3, max(c4)
from table_b
group by c1, c2, c3) table_c
where table_a.c1 = table_c.c1
and table_a.c2 = table_c.c2
and table_a.c3 = table_c.c3
The real query is not simple as above. More tables come
after "the from clause".
Table A and table B are big tables, which have more than
100,000,000 rows.
The response time of this sql is very very slow
as everyone can expect.
Are there any solutions or sql-tips about the late response-time?
I am considering adding a new column into "Table B" in
order to identify the row, which has max serial number.
At this point, I am not sure adding a column is a good
thing in terms of every aspect.
I will be waiting for your advice and every response
will be appreciated even if it is not the solution.
Have a good day.
HO.
Message was edited by:
user507290For such big sources check that
1) you use full scans, hash joins or at least merge joins
2) you scan your source data as less as possible. In the best case each necessary table only once (for example not using exists clause to effectively scan all table via index scan).
3) how much time you are spending on sorts and hash joins (either from v$session_longops directly or some tool that visualises this info). If you are using workarea_size_policy = auto, probably you can switch to manual for this particular select and adjust sort_area_size and hash_area_size big enough to do as less as possible sorts on disk
4) if you have enough free resources i.e. big box probably you can consider using some parallelism
5) if your full scans are taking big time check what is your db_file_multiblock_read_count, probably increasing it for this select will give some gain.
6) run trace and check on what are you waiting for
7) most probably your problem is IO bound so probably you can do something from OS side to make IO faster
8) if your query now is optimized as much as you can, disks are running as mad and you are using all RAM then probably it is the most you can get out of your box :)
9) if nothing helps then you can start thinking about precalculations either using your idea about derived column or some materialized views.
10) I hope you have a test box and at least point (9) do firstly on it and see whether it helps.
Gints Plivna
http://www.gplivna.eu -
ITunes app not loading top tens
The iTunes app on my iPhone4 doesn't load the top tens properly. When I click on More Top Tens it goes back to the main page.
Anyone else found this issue?Actually its fixed...no idea when or how.
-
No of columns in a table and SQL performance
How does the table size effects sql performance?
I am comparing 2 tables , with same number of rows(54 million rows) ,
table1(columns a,b,c,d,e,f..) has 40 columns
table2 (columns (a,b,c,d)
SQL uses columns a,b.
SQL using table2 runs in 1 sec.
SQL using table1 runs in 30 min.
Can any one please let me know how the table size , number of columns in table efects the performance of SQL's?
Thanks
jeevan.user600431 wrote:
This is a general question. I just want to compare table with more columns and table with less columns with same number of rows .
I am finding that table with less columns is good in performance , than the table with more columns.
Assuming there are no row chains , will there be any difference in performance with the number of columns in a table.Jeevan,
the question is not how many columns your table has, but how large your table segment is. If your query runs a full table scan it has to read through the whole table segment, so in that case the size of the table matters.
A table having more columns potentially has a larger row size than a table with less columns but this is not a general rule. Think of large columns, e.g. varchar2 columns, think of blank (NULL) columns and you can easily end up with a table consisting of a single column taking up more space per row than a table with 200 columns consisting only of varchar2(1) columns.
Check the DBA/ALL/USER_SEGMENTS view to determine the size of your two table segments. If you gather statistics on the tables then the dictionary will contain information about the average row size.
If your query is using indexes then the size of the table won't affect the query performance significantly in many cases.
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/ -
On my iPad2, the buttons for Top Ten, Genres and New Releases are missing. How do I get them back.
On my iPad2, the buttons for Top Ten, Genres and New Releases are missing. How do I get them back.
If the menu bar is hidden then press the F10 key or hold down the Alt key, that should make the menu bar appear.
* Go to "View > Toolbars" and check-mark "Menu Bar" with a click if you want to make that permanent
See [[Menu bar is missing]] -
SQL PERFORMANCE SQL ANALYZER of 11g
I am using 11g on Windows 2000, I want run SQL PERFORMANCE ANALYZER to see the impact of parameter change on some sql’s. Currently, I am using it in a test environment, but eventually I want to apply t to production environment.
Let us say I want to see the effect of different values db_file_muntilbock_readcount.
When I run this in my database, will the values changed impact only the session where I am running sol performance analyzer, or will it impact any other sessions, which are accessing the same database instance. I think, it impacts only the session where SQL Performance analyzer is being run, but want to make sure that is the case?
Appreciate your feedback.I think, it impacts only the session where
SQL Performance analyzer is being run, but want to
make sure that is the case?The database instance is part of a larger 'system' which includes a fixed set of physical resources. Your session, and every other session, work within the constraints of those resources. When you change the current SQL statement, you will be moving the balance between those resources.
For example, a disk can only respond to one access request at a time. A memory location can be used for one piece of data at a time. A DB cache buffer can only reflect one block at a time. There are a lot of 'points of serialization'.
Although the major impact should be on the current session, there will be some impact on every other session in the system.
BY the way, there is an 'edit' button available to you for every post you create. As a courtesy, you could edit the title of the duplicate and let us know it is indeed a duplicate - or you could edit that other thread to ask that other question you were going to ask. -
Help needed here please. I am new to this concept and i am working on a tutorial based on SQL performance and security. I have worked my head round this but now i am stuck.
Here is the questions:
1. Analyse possible performance problems, and suggest solutions for each of the following transactions against the database
a) A manager of a project needs to inspect total planned and actual hours spent on a project broken down by activity.
e.g
Project: xxxxxxxxxxxxxx
Activity Code planned actual (to date)
1 20 25
2 30 30
3 40 24
Total 300 200
Note that actual time spent on an activity must be calculated from the WORK UNIT table.
b)On several lists (e.g. list or combo boxes) in the on-line system it is necessary to identify completed, current, or future projects.
2. Security: Justify and implement solutions at the server that meet the following security requirements
(i)Only members of the Corporate Strategy Department (which is an organisation unit) should be able to enter, update and delete data in the project table. All users should be able to read this information.
(ii)Employees should only be able to read information from the project table (excluding the budget) for projects they are assigned to.
(iii)Only the manager of a project should be able to update (insert, update, delete) any non-key information in the project table relating to that project.
Here is the project tables
set echo on
* Changes
* 4.10.00
* manager of employee on a project included in the employee on project table
* activity table now has compound key, based on ID dependence between project
* and activity
drop table org_unit cascade constraints;
drop table project cascade constraints;
drop table employee cascade constraints;
drop table employee_on_project cascade constraints;
drop table employee_on_activity cascade constraints;
drop table activity cascade constraints;
drop table activity_order cascade constraints;
drop table work_unit cascade constraints;
* org_unit
* type - for example in lmu might be FACULTY, or SCHOOL
CREATE TABLE org_unit
ou_id NUMBER(4) CONSTRAINT ou_pk PRIMARY KEY,
ou_name VARCHAR2(40) CONSTRAINT ou_name_uq UNIQUE
CONSTRAINT ou_name_nn NOT NULL,
ou_type VARCHAR2(30) CONSTRAINT ou_type_nn NOT NULL,
ou_parent_org_id NUMBER(4) CONSTRAINT ou_parent_org_unit_fk
REFERENCES org_unit
* project
CREATE TABLE project
proj_id NUMBER(5) CONSTRAINT project_pk PRIMARY KEY,
proj_name VARCHAR2(40) CONSTRAINT proj_name_uq UNIQUE
CONSTRAINT proj_name_nn NOT NULL,
proj_budget NUMBER(8,2) CONSTRAINT proj_budget_nn NOT NULL,
proj_ou_id NUMBER(4) CONSTRAINT proj_ou_fk REFERENCES org_unit,
proj_planned_start_dt DATE,
proj_planned_finish_dt DATE,
proj_actual_start_dt DATE
* employee
CREATE TABLE employee
emp_id NUMBER(6) CONSTRAINT emp_pk PRIMARY KEY,
emp_name VARCHAR2(40) CONSTRAINT emp_name_nn NOT NULL,
emp_hiredate DATE CONSTRAINT emp_hiredate_nn NOT NULL,
ou_id NUMBER(4) CONSTRAINT emp_ou_fk REFERENCES org_unit
* activity
* note each activity is associated with a project
* act_type is the type of the activity, for example ANALYSIS, DESIGN, BUILD,
* USER ACCEPTANCE TESTING ...
* each activity has a people budget , in other words an amount to spend on
* wages
CREATE TABLE activity
act_id NUMBER(6),
act_proj_id NUMBER(5) CONSTRAINT act_proj_fk REFERENCES project
CONSTRAINT act_proj_id_nn NOT NULL,
act_name VARCHAR2(40) CONSTRAINT act_name_nn NOT NULL,
act_type VARCHAR2(30) CONSTRAINT act_type_nn NOT NULL,
act_planned_start_dt DATE,
act_actual_start_dt DATE,
act_planned_end_dt DATE,
act_actual_end_dt DATE,
act_planned_hours number(6) CONSTRAINT act_planned_hours_nn NOT NULL,
act_people_budget NUMBER(8,2) CONSTRAINT act_people_budget_nn NOT NULL,
CONSTRAINT act_pk PRIMARY KEY (act_id, act_proj_id)
* employee on project
* when an employee is assigned to a project, an hourly rate is set
* remember that the persons manager depends on the project they are on
* the implication being that the manager needs to be assigned to the project
* before the 'managed'
CREATE TABLE employee_on_project
ep_emp_id NUMBER(6) CONSTRAINT ep_emp_fk REFERENCES employee,
ep_proj_id NUMBER(5) CONSTRAINT ep_proj_fk REFERENCES project,
ep_hourly_rate NUMBER(5,2) CONSTRAINT ep_hourly_rate_nn NOT NULL,
ep_mgr_emp_id NUMBER(6),
CONSTRAINT ep_pk PRIMARY KEY(ep_emp_id, ep_proj_id),
CONSTRAINT ep_mgr_fk FOREIGN KEY (ep_mgr_emp_id, ep_proj_id) REFERENCES employee_on_project
* employee on activity
* type - for example in lmu might be FACULTY, or SCHOOL
CREATE TABLE employee_on_activity
ea_emp_id NUMBER(6),
ea_proj_id NUMBER(5),
ea_act_id NUMBER(6),
ea_planned_hours NUMBER(3) CONSTRAINT ea_planned_hours_nn NOT NULL,
CONSTRAINT ea_pk PRIMARY KEY(ea_emp_id, ea_proj_id, ea_act_id),
CONSTRAINT ea_act_fk FOREIGN KEY (ea_act_id, ea_proj_id) REFERENCES activity ,
CONSTRAINT ea_ep_fk FOREIGN KEY (ea_emp_id, ea_proj_id) REFERENCES employee_on_project
* activity order
* only need a prior activity. If activity A is followed by activity B then
(B is the prior activity of A)
CREATE TABLE activity_order
ao_act_id NUMBER(6),
ao_proj_id NUMBER(5),
ao_prior_act_id NUMBER(6),
CONSTRAINT ao_pk PRIMARY KEY (ao_act_id, ao_prior_act_id, ao_proj_id),
CONSTRAINT ao_act_fk FOREIGN KEY (ao_act_id, ao_proj_id) REFERENCES activity (act_id, act_proj_id),
CONSTRAINT ao_prior_act_fk FOREIGN KEY (ao_prior_act_id, ao_proj_id) REFERENCES activity (act_id, act_proj_id)
* work unit
* remember that DATE includes time
CREATE TABLE work_unit
wu_emp_id NUMBER(5),
wu_act_id NUMBER(6),
wu_proj_id NUMBER(5),
wu_start_dt DATE CONSTRAINT wu_start_dt_nn NOT NULL,
wu_end_dt DATE CONSTRAINT wu_end_dt_nn NOT NULL,
CONSTRAINT wu_pk PRIMARY KEY (wu_emp_id, wu_proj_id, wu_act_id, wu_start_dt),
CONSTRAINT wu_ea_fk FOREIGN KEY (wu_emp_id, wu_proj_id, wu_act_id)
REFERENCES employee_on_activity( ea_emp_id, ea_proj_id, ea_act_id)
/* enter data */
start ouins
start empins
start projins
start actins
start aoins
start epins
start eains
start wuins
start pmselect
I have the tables containing ouins and the rest. email me on [email protected] if you want to have a look at the tables.Answer to your 2nd question is easy. Create database roles for the various groups of people who are allowed to access or perform various DML actions.
The assign the various users to these groups. The users will be restricted to what the roles are restricted to.
Look up roles if you are not familiar with it.
Maybe you are looking for
-
Hi Experts, I am facing a strange problem when working with BI 7.0, can you please help me out. One user is facing some authorization problem while executing queries, he is using query through BEX. In order to analyse when i copy his user id and when
-
Hi XI Gurus, Can any body plz explain me about the parameters used in RFC Sender Adapter (like RFC Server Parameters: Program ID, Gateway, etc..) Thanks in advance.
-
Is there any table that holds the history of BEx queries
Hi, I just wanted to know if there is any table that keeps the history of changes done to BEx queries?
-
Do you know what happened to some Notes i made on ipad 'Notes'? I turned on the ipad and they have been deleted. I am not connected to the cloud? How do I get them back??
-
Hi All, I just know about these T-Codes. From my testing, RRMX enable me to open Bex Analyzer from SAP GUI, this excel session is already connected to SAP. RRMXP is similar, but this one can open query or workbook directly. Strange thing is that the