Indexing for a simple SQL
Hi guys,.
The below SQL is to do a full table scan.
SELECT AA, STATUS, CC, DD, FF, FF, GG, INCIDENT_ID
OF THE TEST
WHEN STATUS = 'F'
AND INCIDENT_ID = '15434';
1. After you have created an index on two columns STATUS and INCIDENT_ID, he did a scan of interval.
PLAN_TABLE_OUTPUT
----------------------------------------------------------------------------------------------------
---------------------------------------------------
2 - access ("STATUS" = 'F' AND "INCIDENT_ID" = '15434')
2. I have delete the index and create again but this time on a single column INCIDENT_ID. It has the same behavior as the foregoing, do a scan of interval.
PLAN_TABLE_OUTPUT
-------------------------------------------
-------------------------------------------
1 - filter ("DELETE_IND" = 'F')
2 - access ("PAYMENT_DET_ID" = '15434')
The only different is the predicate as above.
Plan with the same cost, explain.
Therefore, can I create indexes only for incident_id? instead for 2 columns? What is actually recommended?
SELECT AA, STATUS, CC, DD, FF, FF, GG, INCIDENT_ID
OF THE TEST
WHEN STATUS = 'F'
AND INCIDENT_ID = '15434';
Thank you
Hello
(1) there seems to be an inconsistency in your message: you use STATUS/INCIDENT_ID in one case and DELETE_IND/PAYMENT_DET_ID in the other, it's confusing
(2) the difference between a single column and several (concatenated) column index here is quite simple. A multi-column index will match your completely predicate, that is, lines that meet the two conditions will return the INDEX RANGE SCAN. A single-column index will only select lines according to the requirement on the indexed column. So the question is, is quite good selectivity?
Let me give you an example: imagine that you have a table that describes the entire population of our planet, and you consider indexing columns COUNTRY_OF_ORIGIN and NATIVE_LANGUAGE. In this case, it is obvious that NATIVE_LANGUAGE does not provide any significant additional selectivity (as in most countries, there are only few languages), so a single-column index would suffice. If you consider a different pair of columns, for example COUNTRY_OF_ORIGIN and AGE, then combined selectivity is much stronger that the selectivity of each column only, so a concatenated index would be useful.
So all down to selectivity, namely, how to combine selectivity refers to the selectivity of single column.
Best regards
Nikolai
Tags: Database
Similar Questions
-
Output is not as expected (getting wrapped) for a simple sql script
Hello
I try to run a simple sql script with a few queries including a semi colon output (;) when text file separated that later - I can open in excel.
My SQL file contains in part the code - below
SELECT
"PORTFOLIO" | « ; » ||
"CONTRACT # | « ; » ||
"CONS_INV #" | « ; » ||
"CUSTOMER #" | « ; » ||
"CLIENT_NAME | « ; » ||
'MESSAGE ' | « ; » ||
"REQUEST_ID" | « ; » ||
"BILL_TO_COUNTRY" | « ; » ||
"BILL_TO_ADDRESS1" | « ; » ||
"BILL_TO_ADDRESS2" | « ; » ||
"BILL_TO_ADDRESS3" | « ; » ||
"BILL_TO_ADDRESS4" | « ; » ||
"BILL_TO_CITY" | « ; » ||
"BILL_TO_POSTAL_CODE" | « ; » ||
"BILL_TO_STATE" | « ; » ||
"BILL_TO_PROVINCE" | « ; » ||
"BILL_TO_COUNTY" | « ; » ||
"SHIP_TO_COUNTRY | « ; » ||
"SHIP_TO_ADDRESS1" | « ; » ||
"SHIP_TO_ADDRESS2" | « ; » ||
"SHIP_TO_ADDRESS3" | « ; » ||
"SHIP_TO_ADDRESS4" | « ; » ||
"SHIP_TO_CITY" | « ; » ||
"SHIP_TO_POSTAL_CODE" | « ; » ||
"SHIP_TO_STATE | « ; » ||
"SHIP_TO_PROVINCE" | « ; » ||
"SHIP_TO_COUNTY" | « ; » ||
"INVOICE_DATE | « ; » ||
"INVOICE_CREATION_DATE" | « ; » ||
"PARTIALLY_DROPPED" | « ; » ||
"AMOUNT".
OF the double
/
SELECT "DOUBLE
/
SELECT "error Code" | « ; » ||' Error message "OF the DOUBLE
/
SELECT lookup_code. « ; » || Description
OF fnd_lookup_values
WHERE lookup_type = 'CUST_ERROR. '
/
The issue I'm facing is that the output of the competitive program file comes out bit so enter each line only 132 characters in a line and therefore gets pushed in the next row in excel.
Output-
PORTFOLIO; CONTRACT #; CONS_INV #; CUSTOMER #; CLIENT_NAME; MESSAGE; REQUEST_ID; BILL_TO_COUNTRY; BILL_TO_ADDRESS1; BILL_TO_ADDRESS2; BILL_TO_A
DDRESS3; BILL_TO_ADDRESS4; BILL_TO_CITY; BILL_TO_POSTAL_CODE; BILL_TO_STATE; BILL_TO_PROVINCE; BILL_TO_COUNTY; SHIP_TO_COUNTRY; SHIP_TO_ADDR
ESS1; SHIP_TO_ADDRESS2; SHIP_TO_ADDRESS3; SHIP_TO_ADDRESS4; SHIP_TO_CITY; SHIP_TO_POSTAL_CODE; SHIP_TO_STATE; SHIP_TO_PROVINCE; SHIP_TO_COUN
TY; INVOICE_DATE; INVOICE_CREATION_DATE; PARTIALLY_DROPPED; AMOUNT
Please advice on all of the commands that I can use in the script sql in order to expand the output page or any other workaround solutions!
Please answer as soon as possible if you can help me!
Thank you!SQL * MORE?
SQL> set lin 1000 SQL> spool your_file_out.csv SQL> your_query SQL> spool offKind regards
Malakshinov Sayan -
All parameter memory index for Oracle text indexes
Hi Experts,
I'm on Oracle 11.2.0.3 on Linux and have implemented Oracle Text. I'm not an expert in this area and need help on a question. I created the Oracle text index with the default setting. However, in a white paper of oracle, I read that the default setting is perhaps not good. Excerpt from the white paper by Roger Ford:
URL:http://www.oracle.com/technetwork/database/enterprise-edition/index-maintenance-089308.html
"(Part of this white paper below. )" ...)
Index memory as mentioned above, $I entries cached emptied out on the disk each time the indexing memory is exhausted. The default index to installing memory is a simple 12MB, which is very low. At the time of creating indexes, users can specify up to 50 MB, but it is still quite low.
This would be by a CREATE INDEX, something like statement:
CREATE INDEX myindex ON mytable(mycol) INDEXTYPE IS ctxsys.context PARAMETERS ('index memory 50M');Allow index of the parameters of memory beyond 50 MB, the CTXSYS user must first of all increase the value of the MAX_INDEX_MEMORY parameter, like this:
begin ctx_adm.set_parameter('max_index_memory', '500M'); end;The parameter memory must never be high point causes paging, because this will have a serious effect on indexing speed. The smallest of dedicated systems, it is sometimes advantageous to temporarily reduce the amount of memory consumed by the Oracle SGA (for example by reducing DB_CACHE_SIZE and/or SHARED_POOL_SIZE) during the index creation process. Once the index has been created, the size of the SGA can be increased again to improve query performance. & quot;
(End of the excerpt from the white paper)
My question is:(1) to apply this procedure (ctx_adm.set_parameter) obliged me to log on as user CTXSYS. Is this fair? or can it be avoided and will be based on the application schema? The CTXSYS user is locked by default and I had to unlock it. Is this OK to do it in production?
(2) what value I should use for the max_index_memory there are 500 MB - my SGA is 2 GB in Dev / QA and / 3 GB in the production. Also in the creation of the index which is the value should I set for the parameter memory index - I had left to default, but how do I change now? Should it be 50 MB as shown in the example above?
(3) the white paper also refer to the reconstruction of an index to an interval like once a month: ALTER INDEX DR$ index_name$ X REBUILD online;
-Is this good advice? I would like to ask the experts once before doing this. We are on Oracle 11 g and the white paper was drafted in 2003.
Basically, while I read the book, I'm still not clear on many aspects and need help to understand this.
Thank you
OrauserN
Index entries are built in memory, and then flushed to disk, memory is exhausted. With a setting of high index memory will mean the index entries can be longer and less fragmented, which provides better performance of the query. Conversely, a small memory parameter index will mean emptied the disk pieces are smaller, so queries (especially on common words) will have to do a lot more e/s to extract several pieces of index for the words.
-
ENQ: TX - index for cars running
Hi all
The system configuration is node 2 11 GR 2 cars used by 11.5.10.2 application ERP on RHEL 5
Face claim to index to insert it on the big table with 27 million.
We have solved this after purging of data to reduce the lines to 4 million and rebuild the index.
I was against the approach to keep mind that the tree index B are self balanced do not rebuold
I have presentation Joel and JLewis comments on this wait event
QuestionsIndex stats before purge : SQL> select NAME,HEIGHT,BLOCKS,LF_ROWS,LF_BLKS ,LF_BLK_LEN,BR_BLKS,DISTI NCT_KEYS ,BTREE_SPACE,USED_SPACE ,PCT_USED from index_stats; NAME HEIGHT BLOCKS LF_ROWS LF_BLKS LF_BLK_LEN BR_BLKS DISTINCT_KEYS BTREE_SPACE USED_SPACE PCT_USED ------------------------------ ---------- ---------- ---------- ---------- ---------- ---------- ------------- ----------- ---------- ---------- OE_PROCESSING_MSGS_TL_U1 3 151040 27108882 149878 4276 422 27108882 644266144 516301179 81 Index stats After purge : SQL> select NAME,HEIGHT,BLOCKS,LF_ROWS,LF_BLKS ,LF_BLK_LEN,BR_BLKS,DISTINCT_KEYS ,BTREE_SPACE,USED_SPACE ,PCT_USED from index_stats; NAME HEIGHT BLOCKS LF_ROWS LF_BLKS LF_BLK_LEN BR_BLKS DISTINCT_KEYS BTREE_SPACE USED_SPACE PCT_USED ------------------------------ ---------- ---------- ---------- ---------- ---------- ---------- ------------- ----------- ---------- ---------- OE_PROCESSING_MSGS_TL_U1 3 14816 4763902 14556 5644 34 4763902 82618912 90654959 110 ASH data before data purge ----------------------- Top User Events DB/Inst: PROD/PROD1 (Apr 04 12:00 to 13:00) Avg Active Event Event Class % Event Sessions ----------------------------------- --------------- ---------- ---------- db file sequential read User I/O 32.36 9.69 read by other session User I/O 19.43 5.82 enq: TX - index contention Concurrency 11.29 3.38 enq: TX - row lock contention Application 10.78 3.23 CPU + Wait for CPU CPU 7.32 2.19 ------------------------------------------------------------- Top Service/Module DB/Inst: PROD/PROD1 (Apr 04 12:00 to 13:00) Service Module % Activity Action % Action -------------- ------------------------ ---------- ------------------ ---------- PROD NPIWIPLR 23.90 Concurrent Request 23.90 OMRSVORD 9.44 Concurrent Request 9.44 NPIIMGION 8.65 Concurrent Request 8.65 OEOIMP 4.53 Concurrent Request 4.53 OEXOEORD 3.83 FRM:B82562:NPI Cus 1.09 ------------------------------------------------------------- Top User Events DB/Inst: PROD/PROD2 (Apr 04 12:00 to 13:00) Avg Active Event Event Class % Event Sessions ----------------------------------- --------------- ---------- ---------- db file sequential read User I/O 30.41 5.62 CPU + Wait for CPU CPU 22.40 4.14 enq: TX - row lock contention Application 13.71 2.53 enq: TX - index contention Concurrency 8.24 1.52 read by other session User I/O 7.80 1.44 ------------------------------------------------------------- Service Module % Activity Action % Action -------------- ------------------------ ---------- ------------------ ---------- PROD NPIWIPLR 10.21 Concurrent Request 10.21 OEXOIMPT 9.60 FRM:B60221:NPI Cus 4.31 FRM:N11217:NPI Cus 2.75 FRM:N13737:NPI Ord 2.54 OEXOEORD 8.78 FRM:N11247:NPI Cus 3.80 FRM:C35809:NPI Cus 1.34 FRM:N11218:NPI Cus 1.20 NPIIMGION 6.67 Concurrent Request 6.67 NPISLIM 5.41 Concurrent Request 5.41 -------------------------------------------------------------
(1) even if it seems he does not index unique sequence but the claim is there?
(2) after reconstruction, we see pct_used > 110, what will happen?
(3) prior use was 81 per cent, is higher and would lead to node splits for large indsddexes?
Concerning
KGKG says:
Face claim to index to insert it on the big table with 27 million.
We have solved this after purging of data to reduce the lines to 4 million and rebuild the index.
I was against the approach to keep mind that the tree index B are self balanced do not rebuoldFirst point to note - if you delete M 23 rows in a table with 27 million lines, it is probably a good idea to rebuild the table and the index to recover the space empty, unless you wait another 20 million lines to be arriviing soon and re-use of space very soon. In fact, you can use the dbms_redefinition package to extract the data you want to keep rather than a deletion. Alternatively you could actually marked as unused as an index on the table that you can get away with before to delete data and (optionally) make an area of narrowing.
>
Index before you purge statistics:
SQL> select NAME,HEIGHT,BLOCKS,LF_ROWS,LF_BLKS ,LF_BLK_LEN,BR_BLKS,DISTI NCT_KEYS ,BTREE_SPACE,USED_SPACE ,PCT_USED from index_stats; NAME HEIGHT BLOCKS LF_ROWS LF_BLKS LF_BLK_LEN BR_BLKS DISTINCT_KEYS BTREE_SPACE USED_SPACE PCT_USED ------------------------------ ---------- ---------- ---------- ---------- ---------- ---------- ------------- ----------- ---------- ---------- OE_PROCESSING_MSGS_TL_U1 3 151040 27108882 149878 4276 422 27108882 644266144 516301179 81 Index stats After purge : SQL> select NAME,HEIGHT,BLOCKS,LF_ROWS,LF_BLKS ,LF_BLK_LEN,BR_BLKS,DISTINCT_KEYS ,BTREE_SPACE,USED_SPACE ,PCT_USED from index_stats; NAME HEIGHT BLOCKS LF_ROWS LF_BLKS LF_BLK_LEN BR_BLKS DISTINCT_KEYS BTREE_SPACE USED_SPACE PCT_USED ------------------------------ ---------- ---------- ---------- ---------- ---------- ---------- ------------- ----------- ---------- ---------- OE_PROCESSING_MSGS_TL_U1 3 14816 4763902 14556 5644 34 4763902 82618912 90654959 110(1) even if it seems he does not index unique sequence but the claim is there?
(2) after reconstruction, we see pct_used > 110, what will happen?
(3) prior use was 81%, it is higher and would lead to node splits large index?One of the comments in one of the blogs I wrote on the index highlights a key problem: http://jonathanlewis.wordpress.com/2009/07/20/index-quiz-1/#comment-33897
Note the LF_BLK_LEN in both cases: it is well below the 8000 bytes you get in general for an index btree, built in a block size of 8 KB. This is because the ITL list in some of the blocks sheet has grown to manage concurrent changes to the blocks, and the command ANALYZE took the biggest list ITL as typical of ALL blocks in the index. 110% and 81% are a side effect of this as Oracle took the used_space and the number of lf_blks * lf_blk_len to calculate the percenage: for example 90654959 / (5644 * 14556) = 1.1035You have created the index with a very large INITRANS, either a number of pads of sheets have already been under pressure concurrent insert and hit the "ITL Explosion" bug, losing a lot of inputs ITL which can not be recovered without some sort of index maintenance.
You said that the index is not based on a sequence - but is it based on some form of monotonous increasing value (for example, a timestamp), you don't need a sequence to hit the problem you just need values that continue to increase. The problem is the number of basis 8767925 bug, fixed in paragraph 12.1 with backports available for some versions.
Concerning
Jonathan Lewis -
Create indexes for an XMLType column.
Hello
I have a table that has a column of XMLType, the XSD is saved in DB. The XML is something like -
< primary >
< sub1 >
< child1_1 / >
< child1_2 / >
< child1_3 / >
< / sub1 >
< sub2 >
< child2_1 / >
< child2_2 / >
< child2_3 / >
< / sub2 >
< sub3 >
< child31_1 / >
< child3_2 / >
< child3_3 / >
< / sub3 >
< / elementary school >
I want to create indexes for child nodes. I read THAT XML structure cannot have collection as index it should be singleton.
So can someone tell me no work around to create indexes for these nodes, or whatever it is that I missed as follows.
Thank you#1. Since you're on 11g stop using the syntax of depricated... (extractValue, table (xmlsequence))
SQL> create table PURCHASEORDER_TAB 2 ( 3 PID number(4), 4 PURCHASEORDER XMLTYPE 5 ) 6 xmltype column PURCHASEORDER 7 store as OBJECT RELATIONAL 8 XMLSCHEMA "http://localhost:80/publishedContent/SB-XML/xsd/2010/purchaseOrder.xsd" Element "PurchaseOrder" 9 / Table created. SQL> select table_name, table_type_name 2 from user_nested_tables where parent_table_name = 'PURCHASEORDER_TAB'; TABLE_NAME TABLE_TYPE_NAME ------------------------------ ------------------------------ SYS_NTdmDOdAAMQsuXdRy6lc0H/A== Action64442_COLL SYS_NT8EIzAbUOQFGN2u6hCXP39g== LineItem64443_COLL SQL> rename "SYS_NT8EIzAbUOQFGN2u6hCXP39g==" to LINEITEM_NT 2 / Table renamed. SQL> create index QTY_INDEX on LINEITEM_NT ("Quantity"); Index created. SQL> set autotrace on explain lines 250 pages 50 trimspool on SQL> -- SQL> select t.pid, quantity 2 from PURCHASEORDER_TAB t, 3 XMLTABLE 4 ( 5 '/PurchaseOrder/LineItems/LineItem' 6 passing t.PURCHASEORDER 7 columns 8 QUANTITY number(4) path 'Quantity' 9 ) 10 where QUANTITY = 3 11 / no rows selected Execution Plan ---------------------------------------------------------- Plan hash value: 1929272390 -------------------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | -------------------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 1 | 46 | 3 (0)| 00:00:01 | | 1 | NESTED LOOPS | | | | | | | 2 | NESTED LOOPS | | 1 | 46 | 3 (0)| 00:00:01 | |* 3 | TABLE ACCESS FULL | LINEITEM_NT | 1 | 23 | 2 (0)| 00:00:01 | |* 4 | INDEX UNIQUE SCAN | SYS_C00429284 | 1 | | 0 (0)| 00:00:01 | | 5 | TABLE ACCESS BY INDEX ROWID| PURCHASEORDER_TAB | 1 | 23 | 1 (0)| 00:00:01 | -------------------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 3 - filter(CAST(TO_NUMBER(TO_CHAR("Quantity")) AS number(4) )=3) 4 - access("NESTED_TABLE_ID"="T"."SYS_NC0002900030$") Note ----- - dynamic sampling used for this statement (level=2) SQL> select t.pid, quantity 2 from PURCHASEORDER_TAB t, 3 XMLTABLE 4 ( 5 '/PurchaseOrder/LineItems/LineItem' 6 passing t.PURCHASEORDER 7 columns 8 QUANTITY number(12,4) path 'Quantity' 9 ) 10 where QUANTITY = 3 11 / no rows selected Execution Plan ---------------------------------------------------------- Plan hash value: 3093486066 -------------------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | -------------------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 1 | 46 | 2 (0)| 00:00:01 | | 1 | NESTED LOOPS | | | | | | | 2 | NESTED LOOPS | | 1 | 46 | 2 (0)| 00:00:01 | | 3 | TABLE ACCESS FULL | PURCHASEORDER_TAB | 1 | 23 | 2 (0)| 00:00:01 | |* 4 | INDEX RANGE SCAN | QTY_INDEX | 1 | | 0 (0)| 00:00:01 | |* 5 | TABLE ACCESS BY INDEX ROWID| LINEITEM_NT | 1 | 23 | 0 (0)| 00:00:01 | -------------------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 4 - access("Quantity"=3) 5 - filter("NESTED_TABLE_ID"="T"."SYS_NC0002900030$") Note ----- - dynamic sampling used for this statement (level=2) SQL> select t.pid, quantity 2 from PURCHASEORDER_TAB t, 3 XMLTABLE 4 ( 5 '/PurchaseOrder/LineItems/LineItem' 6 passing t.PURCHASEORDER 7 columns 8 QUANTITY path 'Quantity' 9 ) 10 where QUANTITY = 3 11 / no rows selected Execution Plan ---------------------------------------------------------- Plan hash value: 3093486066 -------------------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | -------------------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 1 | 46 | 2 (0)| 00:00:01 | | 1 | NESTED LOOPS | | | | | | | 2 | NESTED LOOPS | | 1 | 46 | 2 (0)| 00:00:01 | | 3 | TABLE ACCESS FULL | PURCHASEORDER_TAB | 1 | 23 | 2 (0)| 00:00:01 | |* 4 | INDEX RANGE SCAN | QTY_INDEX | 1 | | 0 (0)| 00:00:01 | |* 5 | TABLE ACCESS BY INDEX ROWID| LINEITEM_NT | 1 | 23 | 0 (0)| 00:00:01 | -------------------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 4 - access("Quantity"=3) 5 - filter("NESTED_TABLE_ID"="T"."SYS_NC0002900030$") Note ----- - dynamic sampling used for this statement (level=2) SQL> SELECT t.pid, extractValue(Value(sta),'/LineItem/Quantity') 2 FROM PURCHASEORDER_TAB t, 3 table(xmlsequence(extract(t.purchaseOrder,'/PurchaseOrder/LineItems/LineItem'))) sta 4 WHERE extractValue(Value(sta),'/LineItem/Quantity')= 3 5 / no rows selected Execution Plan ---------------------------------------------------------- Plan hash value: 3093486066 -------------------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | -------------------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 1 | 46 | 2 (0)| 00:00:01 | | 1 | NESTED LOOPS | | | | | | | 2 | NESTED LOOPS | | 1 | 46 | 2 (0)| 00:00:01 | | 3 | TABLE ACCESS FULL | PURCHASEORDER_TAB | 1 | 23 | 2 (0)| 00:00:01 | |* 4 | INDEX RANGE SCAN | QTY_INDEX | 1 | | 0 (0)| 00:00:01 | |* 5 | TABLE ACCESS BY INDEX ROWID| LINEITEM_NT | 1 | 23 | 0 (0)| 00:00:01 | -------------------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 4 - access("Quantity"=3) 5 - filter("NESTED_TABLE_ID"="T"."SYS_NC0002900030$") Note ----- - dynamic sampling used for this statement (level=2) SQL>Note the first example XMLTable does not use the index because the data type of AMOUNT is given wrong...
-
Read the index for performance data
Hello.
Advice needed.
I have the table as
create table test (a varchar2 (20), ch. varchar2 (20), b varchar2 (20), d...)
I have indexes for the column "a".
Now when I'm data select 1st three columns (a, b, c) by 'a' of the criteria (select a, b, c fom test where a like '% aa') I'll have the execution of plan:
1 scan interval Index
2. access by rowid
Now, I want to choose without 2nd operation. So I'm adding complex index for a, b, c.
After that this select (select a, b, c fom test where a like '% aa') will read the index data.
The question: is it possible to win some perfomance win this way? I mean, it is possible that the research by complex index (a, b, c) will take longer than the search by simple index (a) + access by rowid?
-
How to make this simple sql run faster?
Here is a very simple sql, but it takes too long to finish as an hour or more, and the problem is in the table. It's a huge table with 17660025 lines.
In addition, upd_dt is indexed.
Execution plan seems to be no problem: (is 9i, and I do not think that it would not perform better if the UNIQUE HASH is used instead of SORT UNIQUE)select distinct (dstn_type) from ims_ets_prty_msge where ims_ets_prty_msge.upd_dt > sysdate - 30;
Here is some info on the table----------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost | ----------------------------------------------------------------------------------------- | 0 | SELECT STATEMENT | | 6 | 78 | 1409 | | 1 | SORT UNIQUE | | 6 | 78 | 1409 | | 2 | TABLE ACCESS BY INDEX ROWID| IMS_ETS_PRTY_MSGE | 856K| 10M| 10 | |* 3 | INDEX RANGE SCAN | IMS_ETS_PRTY_MSGE_INDX5 | 154K| | 2 | ----------------------------------------------------------------------------------------- Predicate Information (identified by operation id): --------------------------------------------------- 3 - access("IMS_ETS_PRTY_MSGE"."UPD_DT">SYSDATE@!-30) Note: cpu costing is off
The thing is that we must make it run faster.NUM_ROWS BLOCKS EMPTY_BLOCKS AVG_SPACE CHAIN_CNT AVG_ROW_LEN 17129540 455259 622 502 455278 188
And all I can think is to create a bitmap on dstn_type (low cardinality) index and partition the table using upd_dt;
So my question would be is anyway that we can get a better response time without having to make that kind of change?
Or is this kind of change really help?
Any thoughts would be appreciated!
Thank youHave you thought about putting an index on (upd_dt, dstn_type)?
With the query you posted, and this index, it should be possible to use a full scan of the index, no visit to the table at all.
A bitmap index would not be a good idea, if it is an OLTP application / table.
-
Hello
I have a number of varchar2 fields on my paintings. They are VarChar2 (2000) with an average length between 130 and 290 characters.
I would like to build the index for these fields to improve search performance.
It is on an instance of Oracle9.
Could you recommend a type of index for this field?
Thank you
Death :)Pavan - a DBA wrote:
According to Oracle,.
If this column contains the low cardinality (less), you can create bitmap indexes
If this column contains the high cardinality (more), you can create the index b-tree (default index type)It is not so simple.
Please read the following thread, it contains also some interesting links:
Re: index bitmap and distinct keysNicolas.
-
Is there a way to get a 'default' DataTemplate based on a simple sql query?
Hi all
I wonder if there is a way to produce a default value of the DataTemplate (or at least the dataStructure of the model portion).
I like to call a trigger beforeReport to fill in the data for the report as such, I have to use dataTemplate instead of a simple SQL query, "select * from table_name". I have more than 50 columns on the report so I was hoping that I could not manually built a dataTemplate (or at less the part of dataStructure) for each report just be able to call a trigger.
Thanks in advance for your ideas.
Kind regards
YahyaYou need not define the cases you need the xml structure to be automatically decided by BEEP.
This datatemplate allows you to get the following structure
.. .. .. .. .. .. |
/LIST_ROW>.. .. -
Plans of multiple executions for the same SQL statement
Dear experts,
awrsqrpt. SQL shows several plans for a single SQL statement executions. How is it possible that a single SQL statement will be several Plans of executions within the AWR report.
Here is the output of the awrsqrpt for your reference.
Your contribution is very much appreciated.WORKLOAD REPOSITORY SQL Report Snapshot Period Summary DB Name DB Id Instance Inst Num Release RAC Host ------------ ----------- ------------ -------- ----------- --- ------------ TESTDB 2157605839 TESTDB1 1 10.2.0.3.0 YES testhost1 Snap Id Snap Time Sessions Curs/Sess --------- ------------------- -------- --------- Begin Snap: 32541 11-Oct-08 21:00:13 248 141.1 End Snap: 32542 11-Oct-08 21:15:06 245 143.4 Elapsed: 14.88 (mins) DB Time: 12.18 (mins) SQL Summary DB/Inst: TESTDB/TESTDB1 Snaps: 32541-32542 Elapsed SQL Id Time (ms) ------------- ---------- 51szt7b736bmg 25,131 Module: SQL*Plus UPDATE TEST SET TEST_TRN_DAY_CL = (SELECT (NVL(ACCT_CR_BAL,0) + NVL(ACCT_DR_BAL, 0)) FROM ACCT WHERE ACCT_TRN_DT = (:B1 ) AND TEST_ACC_NB = ACCT_ACC_NB(+)) WHERE TEST_BATCH_DT = (:B1 ) ------------------------------------------------------------- SQL ID: 51szt7b736bmg DB/Inst: TESTDB/TESTDB1 Snaps: 32541-32542 -> 1st Capture and Last Capture Snap IDs refer to Snapshot IDs witin the snapshot range -> UPDATE TEST SET TEST_TRN_DAY_CL = (SELECT (NVL(ACCT_CR_BAL,0) + NVL(AC... Plan Hash Total Elapsed 1st Capture Last Capture # Value Time(ms) Executions Snap ID Snap ID --- ---------------- ---------------- ------------- ------------- -------------- 1 2960830398 25,131 1 32542 32542 2 3834848140 0 0 32542 32542 ------------------------------------------------------------- Plan 1(PHV: 2960830398) ----------------------- Plan Statistics DB/Inst: TESTDB/TESTDB1 Snaps: 32541-32542 -> % Total DB Time is the Elapsed Time of the SQL statement divided into the Total Database Time multiplied by 100 Stat Name Statement Per Execution % Snap ---------------------------------------- ---------- -------------- ------- Elapsed Time (ms) 25,131 25,130.7 3.4 CPU Time (ms) 23,270 23,270.2 3.9 Executions 1 N/A N/A Buffer Gets 2,626,166 2,626,166.0 14.6 Disk Reads 305 305.0 0.3 Parse Calls 1 1.0 0.0 Rows 371,735 371,735.0 N/A User I/O Wait Time (ms) 564 N/A N/A Cluster Wait Time (ms) 0 N/A N/A Application Wait Time (ms) 0 N/A N/A Concurrency Wait Time (ms) 0 N/A N/A Invalidations 0 N/A N/A Version Count 2 N/A N/A Sharable Mem(KB) 26 N/A N/A ------------------------------------------------------------- Execution Plan ------------------------------------------------------------------------------------------------ | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | ------------------------------------------------------------------------------------------------ | 0 | UPDATE STATEMENT | | | | 1110 (100)| | | 1 | UPDATE | TEST | | | | | | 2 | TABLE ACCESS FULL | TEST | 116K| 2740K| 1110 (2)| 00:00:14 | | 3 | TABLE ACCESS BY INDEX ROWID| ACCT | 1 | 26 | 5 (0)| 00:00:01 | | 4 | INDEX RANGE SCAN | ACCT_DT_ACC_IDX | 1 | | 4 (0)| 00:00:01 | ------------------------------------------------------------------------------------------------ Plan 2(PHV: 3834848140) ----------------------- Plan Statistics DB/Inst: TESTDB/TESTDB1 Snaps: 32541-32542 -> % Total DB Time is the Elapsed Time of the SQL statement divided into the Total Database Time multiplied by 100 Stat Name Statement Per Execution % Snap ---------------------------------------- ---------- -------------- ------- Elapsed Time (ms) 0 N/A 0.0 CPU Time (ms) 0 N/A 0.0 Executions 0 N/A N/A Buffer Gets 0 N/A 0.0 Disk Reads 0 N/A 0.0 Parse Calls 0 N/A 0.0 Rows 0 N/A N/A User I/O Wait Time (ms) 0 N/A N/A Cluster Wait Time (ms) 0 N/A N/A Application Wait Time (ms) 0 N/A N/A Concurrency Wait Time (ms) 0 N/A N/A Invalidations 0 N/A N/A Version Count 2 N/A N/A Sharable Mem(KB) 26 N/A N/A ------------------------------------------------------------- Execution Plan --------------------------------------------------------------------------------------------- | Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | --------------------------------------------------------------------------------------------- | 0 | UPDATE STATEMENT | | | | 2 (100)| | | 1 | UPDATE | TEST | | | | | | 2 | TABLE ACCESS BY INDEX ROWID| TEST | 1 | 28 | 2 (0)| 00:00:01 | | 3 | INDEX RANGE SCAN | TEST_DT_IND | 1 | | 1 (0)| 00:00:01 | | 4 | TABLE ACCESS BY INDEX ROWID| ACCT | 1 | 26 | 4 (0)| 00:00:01 | | 5 | INDEX RANGE SCAN | INDX_ACCT_DT | 1 | | 3 (0)| 00:00:01 | --------------------------------------------------------------------------------------------- Full SQL Text SQL ID SQL Text ------------ ----------------------------------------------------------------- 51szt7b736bm UPDATE TEST SET TEST_TRN_DAY_CL = (SELECT (NVL(ACCT_CR_BAL, 0) + NVL(ACCT_DR_BAL, 0)) FROM ACCT WHERE ACCT_TRN_DT = (:B1 ) AND PB RN_ACC_NB = ACCT_ACC_NB(+)) WHERE TEST_BATCH_DT = (:B1 )
Thank you for taking your time to answer my question.
ConcerningOracle Lover3 wrote:
How will I know (from Plan 1 and Plan 2) whose execution plan chose for the current run?Since you're already on 10.2, you can identify the actual execution plan by checking in V$ SESSION SQL_ID and SQL_CHILD_NUMBER column. This can be used to identify the plan in V$ SQL_PLAN (columns SQL_ID and CHILD_NUMBER) and in 10g, you can use the convenient DBMS_XPLAN. Function DISPLAY_CURSOR for the information of the real plan using these two parameters.
Kind regards
RandolfOracle related blog stuff:
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/ -
Why is there not an option to create a custom for a simple telephone touch label?
Why is there not an option to create a custom for a simple telephone touch label? 9. new version of iOS.
There are. What makes you think there isn't. When you click to add a phone number, click label, and then scroll down to Custom. Type away.
-
Driver missing for PCI Simple Communications Controller for U41
I have a problem with a driver for PCI Simple Communications Controller for U41. Please help for the best solution thanks
Good day and welcome to the community.
This is usually resolved by installing the Intel Management Engine Interface package.
Please take a look at the pages of resources of your machine, under the section of Chipset :
http://support.Lenovo.com/us/en/products/laptops-and-netbooks/u-series/U41-70
I don't know if there are also the Serial i/o package.
I hope this helps a little.
Kind regards.
-
You are looking for a SIMPLE, EASY to USE, USB flash drive password protection
Hello everyone,
I scoured the Internet looking for a SIMPLE, EASY to USE, method of Word to PASS-PROTECT my flash drives TOGETHER. I don't mean something that protects just individual files & folders (such as MS Word, Excel & can do). Or I'm looking for a program that involves having to read page after page of instructions on how to ENCRYPT the drive. ENCRYPTION is a topic that is really beyond my ability to understand. I'm sorry.
Programs like Truecrypt, Axcrypt, etc. are simply much too complicated for me to understand... a lot less!
My question: are there flash drives available (size 8-16 GB preferably) who have a means of securing the disk with a password?
The USB key remains in my USB port at ALL times! I have no other computers here and live alone, so that no one else could access the flash player (or computer). Otherwise, I would be willing to consider a VERY SIMPLE, EASY to UNDERSTAND AND USE application, either free or something that does not cost me dearly. (I subsist on a limited monthly disability income).
Thank you for your time and any recommendations!
It seems that the solution for you would be to purchase a secure Flash drive. Security is built-in and it asks for your password when you insert your disc into the computer. Can't get much simpler than that. They are a little more expensive, but you pay more for practical reasons. See the following article:
'USB Flash Drive Security '.
<>http://en.Wikipedia.org/wiki/USB_flash_drive_security >
To find one, just Google for "Secure USB flash drive", there are a lot of them out there.If you have a non-secure flash drive and do not want to replace, then your easiest option would be a program like Truecrypt which you already have reduced as being too complicated. Don't let the word"encryption" you scared - it all comes down to "enter a password to access".» Once established, however, it is almost as simple as only to enter a password so if you can find a computer or a local high school computer nerd literate member of the family who would volunteer to set up for you, then it wouldn't be that hard.
HTH,
JW -
How to print an index for a directory?
original title: print the directory
How to print an index for a directory? At home I run Windows 7, but I really want to do this at work - OS is Vista Business
You mean "how to print a list of the directories?
http://www.Google.com/search?q=how+do+I+print+a+directory+listingHow to add the feature print directory for files in Windows XP, Windows Vista, in Windows 7
http://support.Microsoft.com/kb/321379HTH
-
Parse Hard & Soft Parse for a given SQL?
Hi all
I need to know which of these is a hard and soft analysis for a given Sql. What I think is the 'charges' and 'invalidation' indicates if a cursor is invalidated and gets reloaded, which allows us to check if a call analysis becomes hard analysis.
Confusion here is should "Disk reads" value nonzero means hard analysis regardless of the 'load' and 'invalidation' and 'Disc bed' zero value means soft parse?
SELECT a.SQL_ID, a.INSTANCE_NUMBER, a.SNAP_ID, TO_CHAR(b.END_INTERVAL_TIME, 'DD-MON-YYYY HH24:MI') snap_time, a.PLAN_HASH_VALUE, ROUND(a.ELAPSED_TIME_DELTA/1000000) TOTAL_EXECUTION_TIME_SEC, ROUND(a.CPU_TIME_DELTA/1000000) cpu_seconds, a.ROWS_PROCESSED_DELTA rows_processed, a.BUFFER_GETS_DELTA buffer_gets, a.DISK_READS_DELTA disk_reads, a.EXECUTIONS_DELTA executions, a.PARSE_CALLS_DELTA parses, a.VERSION_COUNT version_count, a.LOADS_TOTAL loads_total, a.INVALIDATIONS_TOTAL invalidations_total FROM DBA_HIST_SQLSTAT a, DBA_HIST_SNAPSHOT b WHERE a.SNAP_ID = b.SNAP_ID and a.INSTANCE_NUMBER=b.INSTANCE_NUMBER AND a.INSTANCE_NUMBER=1 AND a.SQL_ID='bsraj0u9r9d3w' ORDER BY a.SNAP_ID desc; TOTAL Instance Snap Snap EXECUTION CPU Buffer Disk version loads invalidations Sql Id Number Id Time TIME(Sec) Seconds Rows Gets Reads Executions Parses count total total ------------- -------- --------- -------------------- ------------ ------- -------- ---------------- -------- ---------- ------ ------- ----- ------------- bsraj0u9r9d3w 1 9368 12-FEB-2016 12:00 14 13 200 14,079,211 780 2 1 1 6 5 --- hard Parse bsraj0u9r9d3w 1 9348 11-FEB-2016 16:00 110 91 1,500 104,806,435 0 15 4 1 5 4 --- ?? bsraj0u9r9d3w 1 9347 11-FEB-2016 15:00 30 29 500 34,922,061 0 5 2 1 5 4 --- ?? bsraj0u9r9d3w 1 9346 11-FEB-2016 14:00 14 13 200 14,009,248 0 2 1 1 5 4 --- ?? bsraj0u9r9d3w 1 9341 11-FEB-2016 09:00 14 13 200 14,009,957 651 2 2 1 5 4 --- hard Parse bsraj0u9r9d3w 1 9297 09-FEB-2016 13:00 20 20 500 23,380,631 0 5 4 1 4 3 --- ?? bsraj0u9r9d3w 1 9294 09-FEB-2016 10:00 65 53 1,300 58,569,220 0 13 13 1 3 2 --- ?? bsraj0u9r9d3w 1 9293 09-FEB-2016 09:00 18 17 300 18,635,057 0 3 2 1 3 2 --- ?? bsraj0u9r9d3w 1 9292 09-FEB-2016 08:00 70 67 1,700 77,145,560 787 17 17 1 3 2 --- hard Parse bsraj0u9r9d3w 1 9272 08-FEB-2016 12:00 18 18 400 22,164,178 0 4 4 1 2 1 --- ?? bsraj0u9r9d3w 1 9270 08-FEB-2016 10:00 106 101 2,500 123,154,635 0 26 24 1 2 1 --- ?? bsraj0u9r9d3w 1 9269 08-FEB-2016 09:00 11 10 300 11,401,899 657 3 3 1 2 1 --- hard Parse bsraj0u9r9d3w 1 9253 07-FEB-2016 17:00 104 98 2,300 115,320,927 788 23 23 1 1 0 --- hard Parse
Thanks in advance
Reads disk do not mean that the query will invalidation thus analyze hard. Reading disc mean simply that the IO that happens from disk and it is perhaps due to the result not being not not available in memory.
I'm not sure why you use this type of analysis to check if requests are going for hard analysis or not? Why not pick up a particular query run and use TKPROF(better SQLTXPLAN) or AWR based SQRPT.sql to check the same? What is the exact issuethat, you're trying to solve?
Aman...
Maybe you are looking for
-
MY button "List all tabs" disappeared, how do I get that back?
I used to have a small arrow in the top right corner of my screen called 'list all tabs' when I clicked it, it gave me a drop down of all my open tabs. Somehow he disappeared and I do not know how to retrieveI tried to go in customize, but could not
-
I can't attach a large excel file with firefox. in an e-mail.
I often attach files to send, but this file of particles will not focus.
-
It seems that the recovery for the M200 DVDs is not complete. I can't start it using an external DVD so followed the procedure of the Portege 3500 in support for http://tinyurl.com/7465q newsletter It started from a floppy and allowed me to view the
-
Bad screen refresh rate shown under Vista - Satellite P100-429
Hello I have a Satellite P100-429. There are three weak, I install Windows Vista Ultimate Edition, but now the highest of screen refresh rate is 60 Hz. I tried to download the latest drivers for the graphics card, but this will not work and the compu
-
DPP 4 cut and reframe the Question to the tool
When using DPP 4 setting / Crop tool I can't paste changes to multiple images. When you use the old version of DPP 3, there is a button "apply to all". It was a very useful feature. I hope that it has not been removed in the latest version of the sof