I have a huge table (about 60 gb) partition over range. The index on this table is global index created on 4 columns together. I have a query which is running very slowly. The explain plan is showing the use of this global index.Explain plan is not showing pstart and pend because the index is global.
View 6 RepliesI have a partioned (by row_create_date) table, lets called it TABLE_X, which has about 300 million records. This table has 7 columns including the primary key and a non-unique, locally partitioned column called trace_id; 99% of queries access this table via this column.
Lately, querying TABLE_X via the trace_id has been very very bad. Queries run for over 1 hr in some cases. So we decided to change the index for trace_id to a global index. Now queries against TABLE_X return in seconds. So far so good.
However, when the query has to join TABLE_X to another table, the query sometimes runs for over 1 hours; back to the same old problem. Here is an illustration;
SELECT COUNT(1) FROM TABLE_X WHERE TRACE_ID = 'XXXXX';
-- returns in seconds
SELECT COUNT(1)
FROM TABLE_X,
TABLE_Y
WHERE TABLE_X.TRACE_ID = 'XXXXX'
AND TABLE_X.TRACE_D = TABLE_Y.TRACE_ID;
-- runs for over 1 hr, even when TABLE_Y.TRACE_ID is a primary key.
I have a global index and I want to convert it to local index.Is there a way to recreate local index with out dropping the global index.
I can create a local index first and then drop the global index. But is there a way to create it with out dropping the global index, just convert it.
I have table with 4 partition by range partition. I am loading the table in bulk mode to latest partition. Before I load , I dropped the index and after Load I will be creating index. So when I am dropping index, it is dropping index from all the partitions and when creating the index, I am creating the index for all partitions. When I am creating index using local, it is telling you have to create local index for all partitions at the same time. because of that I have to drop and recreate all indexes again. Again I have to gather stats for whole table .
I was thinking we can build index for one partition and index should remain as is for old partitions If this is not the case, how do I plan my load for a partitioned table using bulk mode to latest partition.
I was comparing cost of rebuild vs create index...I carried out the following test
SQL> create table t4 as select * from t1;
Table created.
SQL> create table t5 as select * from t1 where 1=2;
Table created.
SQL> create index i5 on t5(id);
Index created. SQL> select bytes,extents,blocks from user_segments where segment_name='I5';
BYTES EXTENTS BLOCKS
---------- ---------- ----------
65536 1 8
SQL> alter index i5 unusable;
Index altered.
SQL> alter table t5 nologging;
Table altered.
SQL> Alter session set skip_unusable_indexes=True;
Session altered.
SQL> insert /*+ append */ into t5 select * from t1;
563904 rows created.
SQL> commit;
Commit complete.
Now I compared the cost (elapsed time, logical I/O) of the operations
create index i4 on t4(id);
Vs
alter index i5 rebuild online;
Following is the related trace of above 2 steps
create index i4 on t4(id)
call count cpu elapsed disk query current rows
------- ------ -------- ---------- ---------- ---------- ---------- ----------
Parse 1 0.00 0.00 0 1 0 0
Execute 1 1.17 3.38 9497 7869 335 0
Fetch 0 0.00 0.00 0 0 0 0
------- ------ -------- ---------- ---------- ---------- ---------- ----------
total 2 1.17 3.38 9497 7870 335 0
Misses in library cache during parse: 1
Optimizer goal: CHOOSE
Parsing user id: 5
[code]....
So which option we shall pick in such cases? {Of course I haven't set 'nologging' for the indices but it is same for both indices we are comparing}
If we have not set parallel degree for a table then we can ( try to ) force parallel execution on a table using a parallel hint Does this 'parallelism' works on the index search in the query as well?
In which situations non-parallel non-partitioned table but parallel index (degree>2) will facilitate a query?
How does standard vs advanced handle index compression? Is there an advantage to having one vs the other besides the incurred cost?
View 3 Replies View RelatedHow to force an index if the table not using the index?
View 10 Replies View Relatedgetting how many local and global indexes on particular oracle table
View 2 Replies View Related2day i was dropping few unwanted index from the data base, By mistake i removed the local partitioned index , So i want to recreate that index.i create the index, will the partitioned index updates when we add partitioned to the tables.
View 2 Replies View RelatedWhat is the use of Local and Global Partition Index?
View 1 Replies View RelatedI'm trying to split a table partition.
whether the below syntax are correct.
If local index used
ALTER TABLE SNYT.PART_ESTD
SPLIT PARTITION ESTD_M13_S22 AT ('ESTD', 13, '22')
INTO (PARTITION ESTD_M13_S21, PARTITION ESTD_M13_S22)
update indexes;
(per http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:1401247200346349807)
=================================================================
If Global index used
ALTER TABLE SNYT.PART_ESTD
SPLIT PARTITION ESTD_M13_S22 AT ('ESTD', 13, '22')
INTO (PARTITION ESTD_M13_S21, PARTITION ESTD_M13_S22)
UPDATE GLOBAL INDEXES;
Version Info: 11.2 on Solaris 10
I have a partitioned table like below. I want to create a B-Tree index on SALES_RGN column which is neither the part of Primary key or the Partitioned key. Should I create this index as local or Global ?
CREATE TABLE sales_dtl
(
txn_id number (9),
salesman_id number(5),
salesman_name varchar2(30),
sales_rgn varchar2(10), -----------------------------> This column needs to be indexed
sales_amount number(10),
sales_date date,
constraint pk_sales_dtl primary key (txn_id)
[code]....
I have a query which had a join:
a.c1=b.c1 and a.c2=@var
where @var is user supplied input at runtime...We had a index on a.c2 . The CBO would use this index to generate an opitimised query plan.We found some records from table "b" were dropping due to inner join. So we made a change in join. It'd be like
a.c1(+)=b.c1 and nvl(a.c2,@var)=@var
This query is no longer using the index, instead its doing a full table scan causing the query to slowdown.I have tried creating index on nvl(a.c2,'31-dec-9999')
But the CBO won't use it.Anyway to create index on this col so that full table scan can be avoided?
Let's consider such table that all rows fit into single block:
SQL> create table test as select rownum id, '$'||rownum name from dual connect by level <= 530;
Table created.
SQL> create index i_test on test(id);
Index created.
SQL>
SQL> begin
[code].....
why does approach with full scan take longer even if table occupies only one data block? PS. 11gR2
16:28:32 SQL> create bitmap index bp_idx_ag_id on transactions(type);
create bitmap index bp_idx_ag_id on transactions(type)
*
ERROR at line 1:ORA-25122: Only LOCAL bitmap indexes are permitted on partitioned tables
how to create bitmap index on partitiioned tables
I created table Rang pertitionned, and List subpartitionned. My table is Interval partitionning. My subpartition is template based as is :
PARTITION BY RANGE ( DINFOIDENTITE )
INTERVAL ( (NUMTOYMINTERVAL(1,'MONTH')) )
SUBPARTITION BY LIST ( AVANT_DERNIER_MATCLE )
[Code]....
i would have the same repartition with 10 differents tablespaces, one for each subpartition.
I search on Oracle documentation this morning, but the only thing i've found is that this option is possible with the STORE IN clause, but only for HASH partitionning.
Is there a way to specify STORAGE clause with a template for my indexes ?
What is the difference between index rebuild and index rebuild online.
View 3 Replies View RelatedI have created an non unique index lk_fein on lookup_fein( code,map_id,trash). When I check the explain plan it does a full table scan on lookup_fein. if I force it to use index by it does and the cost also decreases.
SQL> SELECT WORK_FEIN,
2 NON_FEIN ,
3 FI_FEIN ,
4 MFEIN ,
5 TOTAL_FEIN ,
[code]...
Facts:
Oracle 10.1.0.5
AIX 5.3
My table, HMTX have 10 partitions each of one have 6 millions of rows (average). We have 7 partitioned LOCAL indexes in that table.
Every month we load data into a new partition (6 million of rows aprox) and drop the oldest partition in table HMTX.
In order to do that we have a script that contain the next statements:
drop of all indexes
drop index n1;
drop index n...;
drop index n7;
[Code]...
create indexes again with tha same storage and degree parameters
CREATE INDEX hmtx_TST_N1 ON hmtx (campo1, campo2, campo3 .... campo8)
TABLESPACE xxxx
PCTFREE 0
INITRANS 2
[Code]....
My problem is in rhe index creation section, despite use parallel with degree 8 and nologging the index was created in :
Elapsed: 02:43:50.85.
In past months that index was created in :
Elapsed: 01:43:36.94
Elapsed: 04:48:31.24
Elapsed: 00:57:16.28
there are another way in order speed the index creation ?? o another way to disable ths index ??
How the length of column width effects index performance?
For example if i had IOT table emp_iot with columns:
(id number,
job varchar2(20),
time date,
plan number)
Table key consist of(id, job, time)
Column JOB has fixed list of distinct values ('ANALYST', 'NIGHT_WORKED', etc...).
What performance increase i could expect if in column "job" i would store not names but concrete numbers identifying job names.
For e.g. i would store "1" instead 'ANALYST' and "2" instead 'NIGHT_WORKED'.
I am rebuilding some UNUSABLE local index partitions on Oracle 8.1.7.4.0 (64bit) database . The platform is a HPUX machine.
The DDL of the partition table/indexes:
=========================
CREATE TABLE TESTME
( INST_NO CHAR(3) NOT NULL,
ACCT_NO CHAR(16) NOT NULL,
REC_NO CHAR(9) NOT NULL,
TRAN_TYPE CHAR(2) DEFAULT ' ',
STAT CHAR(2) DEFAULT ' ',
[code]...
We have a DELETE statement when coming from application is not using index but when run from Toad or SQLplus as same user uses index. Explain plan also shows using index.I did a query on v$sql below is the output of the query( I have attached the same as a txt file). All the stats are up to date and confirmed from the developer the variable B1 is using the same datatype as column MAXMKY.
SQL_TEXTSQL_ID DISK_READSOPTIMIZER_HASH_VALUE
DELETE LOTA WHERE MAXMKY=:B1 2g2prrp3z56ah19,099,1891,846,735,884
DELETE LOTA WHERE MAXMKY=:B1 2g2prrp3z56ah0 1,846,735,884
OPTIMIZER_COST HASH_VALUEPLAN_HASH_VALUE MODULEPARSING_SCHEMA_NAME
[code].....
I am working on a query for a feedback response system which is going to be targeted at the common case when the user only want the most recent 10-20 rows in the feedback table. My though is to create an index on the date column, do a sort in an inner query and rownum <= in an outer query. This works as I expect when I am only querying the main table (lookup by index with a stop key), but when I start joining the main table to attribute tables I end up with a full table scan of the main table with the stop key applied after all the joins are completed, the index is nowhere to be found.
CREATE TABLE attr1_tbl(attr1_id NUMBER NOT NULL, attr1 VARCHAR2(10) NOT NULL,
CONSTRAINT attr1_pk PRIMARY KEY (attr1_id));
CREATE TABLE attr2_tbl(attr2_id NUMBER NOT NULL, attr2 VARCHAR2(10) NOT NULL,
CONSTRAINT attr2_pk PRIMARY KEY (attr2_id));
CREATE TABLE attr3_tbl(attr3_id NUMBER NOT NULL, attr3 VARCHAR2(10) NOT NULL,
CONSTRAINT attr3_pk PRIMARY KEY (attr3_id));
[code]....
One thing I noticed was that when no data is selected from the attribute tables, even if they are joined in the query, the CBO throws them out of the plan and only accesses the main table. With the foreign keys this makes sense and really just disqualified my first thought that maybe I was missing a foreign key or not null constraint somewhere.
I also added the cardinality hint to overcome the chance that in my test case there was so little data that index access is not worth it.
I am going through this scenario:
* 35 | ID TABLE ACCESS BY INDEX ROW | S_ORG_EXT | 3064K| 2472M| | 1 (0)| 00:00:01 |
| 36 | INDEX FULL SCAN | S_ORG_EXT_U1 | 14 | | | 1 (0)| 00:00:01 |
Predicate Information (identified by operation id):
---------------------------------------------------
35 - filter("T2"."ACCNT_FLG"<>'N' AND ("T2"."INT_ORG_FLG"<>'Y' OR "T2"."PRTNR_FLG"<>'N'))
This unselective index scan on step 36 of the explain is returning 14 rows but optimizer is selecting 3064 K rows from the table .
I tried creating combined index on all 3 columns mentioned in the predicates for 35th step , but that is not utilized .
how to index this whole expression ::--
(ACCNT_FLG<>'N' AND (INT_ORG_FLG<>'Y' OR PRTNR_FLG<>'N'))
Something like CREATE INDEX XYZ on table((ACCNT_FLG<>'N' AND (INT_ORG_FLG<>'Y' OR PRTNR_FLG<>'N')) compute statistics ;
is there any way to reduce the index creation time.
in my case one index creation took 5 minute and there are 5 indexes , so it took 25 minutes.
I have the following problem. When I used in the IN-Statement fixed values e.q. 197321,197322,197323 ..., the index i_tab2_index works fine (index range scan).
But when I used in the IN-Statement an Sub-Select, the index i_tab2_index doesn't work (fast full scan)!My scale indices and used Selects:
CREATE INDEX i_tab1_index ON tab1 ( datum, flag_inst );
CREATE INDEX i_tab2_index ON tab2 ( tab2Idx, kontro );
SELECT count(epidx) as rowAnz
FROM tab2
WHERE tab2Idx IN ( SELECT tab1IDX FROM tab1
WHERE datum BETWEEN '20120117' AND '20120117'
AND flag_inst = '1' )
AND kontro = '9876521'
[code]...
get all the unused index in the system , if i put this query in batch job and execute it every night upto one months and store its data in a table and after one months i can get all the used indexes and left would be our unused indexes.
select
distinct p.object_name c1
from
dba_hist_sql_plan p,
dba_hist_sqlstat s
[Code]....
I have a table whose size is 2.3 GB and there are two indexes on it. One index is based on a Date column whose size is 900 MB, and the Other index consists of 5 columns including the date column, and the size is almost 2GB. But when i query the table using the Date column, it is doing a range scan on the second index which is almost the same size as the table. why is it not using the first index? What steps should i take so that it uses the First index without passing hints.
View 4 Replies View Relatedmbr has 60,000 rows and member has 60,000 rows approx. two tables have indexes on ssn, and citi_no on them.
PK of mbr : mbr_id
PK of member : mbr_id
other columns are not PK, and have no index on it.
I'm wondering why the statment doesn't use index while ssn and citi_no have index.
MERGE INTO mbr t
USING (SELECT mbr_id,citi_no
FROM member) a
ON (t.ssn = a.citi_no)
WHEN MATCHED THEN
UPDATE SET t.asis_mbr_id = a.mbr_id
where t.ssn not in(select ssn from mbr group by ssn having count(*) > 1)