I am having a problem with a simple query of a view. The view itself is nasty having union_all and most of its fields evaluated by functions. The problem is that even a very simple query can get the whole view evaluated which then takes forever.
To simplify my problem I have made this two queries
select * from horrible_view where id in (1, 2, 3); -- this works super quick
select * from horrible_view where id in (select id from my_temp_table); -- this takes forever
select * from my_temp_table t left join horrible_view h on t.id=h.id -- this also takes forever
So even though my_temp_table contains the same 1, 2, 3 (3 rows with just one column) the second and third query makes the optimizer go bananas, so that the horrible view is evaluated completly and then returns just the 3 rows are returned.
Is there anything I could do?
IE in the second example make optimizer run the "select id from my_temp_table" first and then the query.
Oracle's version is 11.2
Further info:
First off I know only a little about Oracle. (So is the Explain plan from sqlDeveloper the thing you want? If so how can you print it, its huge?) Well here is the screenshot: here
I have created the temp table just in order to test. Its actually named TMP_R (and it has 10 rows)
Related
I want to execute a select statement based on a result set from a previous step, something like this:
select column from table where column in (previous step);
Basically this step (filter rows) will split a group of ids based on a condition. I want to make a select with those who tested false but i don't know how to select only those. The table in question, which I want to select, it's very big and it is very expensive to select all records and join with the result set, so I wish to select just the group that I need, is this even possible ?
https://i.stack.imgur.com/Xu1qt.png
Ok, let me try to be more specific.
Basically I Have 3 steps as my print shows.
First step is a table input, which I select from a table.
Second step is a database lookup, which I look on other table to get some fields that I want.
And third step it's a filter rows, where I kinda make a if else statement.
After my third step (filter rows) I have 2 streams: True or False.
Each stream returns me a group of ids and other fields too but it's not that important here, I guess.
I want to make a select statement based on those ids returned from previous step (Filter rows 3° step).
Basically the behaviour that I want its similar to this query:
select *
from table
where id in ("previous step");
Where table will always be the same table, so I don't think this will be a problem or something.
And "previous step" means all ids returned after the 3° step (filter rows).
What i am doing right now is: I have another table input on the other side which I make a merge join with this result set(from 3° step). But I have to make a select of the entire table and then, join with my result set, what is very expensive, and I'm wondering if i can get the same result, but with more performance.
I don't know if I am being clear enough, but I apologize right now because english it's not my main language, but I hope you guys can understand me now, thanks.
You can use three steps to achive this.
First, use a Memory group by step to group ids as a field.The aggregate tyoe should beConcatenate strings separated by ,
Second,use a User defined java expression step to generate a new field contains the SQL we need.The expression may like"SELECT id,created FROM test WHERE order_id IN ("+ ids +")" and ids is the group result from last step.
At last,we can use a Dynamic SQL row step to look up datas by the specified SQL.
I have a quite complicated view (using several layers of views across several database links) which takes a second to return all of it's rows. But, when I ask for distinct rows, it takes considerably more time. I stopped waiting after 4 minutes.
To make my self as clear as possible:
select a, b from compicated_view; -- takes 1 sec (returns 6 rows)
select distinct a, b from compicated_view; -- takes at least 4 minutes
I find that pretty weird, but hey, that's how it is. I guess Oracle messed something up when planing that query. Now, is there a way to force Oracle to first finish the select without distinct, and then do a "select distinct *" on the results? I looked into optimizer hints, but I can't find anything about hinting the order in which distinct is applied (this is first time I'm optimizing a query, obviously :-/).
I'm using Oracle SQl Developer on Oracle 10g EE.
Try:
SELECT DISTINCT A,B FROM (
SELECT A,B FROM COMPLICATED_VIEW
WHERE rownum > 0 );
this forces to materialize the subquery and prevents from view merging/predicate pushing, and likely from changing the original plan of the view.
You may also try NO_MERGE hint:
SELECT /*+ NO_MERGE(alias) */
DISTINCT a,b
FROM (
SELECT a,b FROM COMPLICATED_VIEW
) alias
Since you haven't posted details... try the following:
SELECT DISTINCT A,B
FROM
(SELECT A,B FROM COMPLICATED_VIEW);
I want a query that selects the number of rows in each table
but they are NOT updated statistically .So such query will not be accurate:
select table_name, num_rows from user_tables
i want to select several schema and each schema has minimum 500 table some of them contain a lot of columns . it will took for me days if i want to update them .
from the site ask tom he suggest a function includes this query
'select count(*)
from ' || p_tname INTO l_columnValue;
such query with count(*) is really slow and it will not give me fast results.
Is there a query that can give me how many rows are in table in a fast way ?
You said in a comment that you want to delete (drop?) empty tables. If you don't want an exact count but only want to know if a table is empty you can do a shortcut count:
select count(*) from table_name where rownum < 2;
The optimiser will stop when it reaches the first row - the execution plan shows a 'count stopkey' operation - so it will be fast. It will return zero for an empty table, and one for a table with any data - you have no idea how much data, but you don't seem to care.
You still have a slight race condition between the count and the drop, of course.
This seems like a very odd thing to want to do - either your application uses the table, in which case dropping it will break something even if it's empty; or it doesn't, in which case it shouldn't matter whether it has (presumably redundant) and it can be dropped regardless. If you think there might be confusion, that sounds like your source (including DDL) control needs some work, maybe?
To check if either table in two schemas have a row, just count from both of them; either with a union:
select max(c) from (
select count(*) as c from schema1.table_name where rownum < 2
union all
select count(*) as c from schema2.table_name where rownum < 2
);
... or with greatest and two sub-selects, e.g.:
select greatest(
(select count(*) from schema1.table_name where rownum < 2),
(select count(*) from schema2.table_name where rownum < 2)
) from dual;
Either would return one if either table has any rows, and would only return zero f they were both empty.
Full Disclosure: I had originally suggested a query that specifically counts a column that's (a) indexed and (b) not null. #AlexPoole and #JustinCave pointed out (please see their comments below) that Oracle will optimize a COUNT(*) to do this anyway. As such, this answer has been altered significantly.
There's a good explanation here for why User_Tables shouldn't be used for accurate row counts, even when statistics are up to date.
If your tables have indexes which can be used to speed up the count by doing an index scan rather than a table scan, Oracle will use them. This will make the counts faster, though not by any means instantaneous. That said, this is the only way I know to get an accurate count.
To check for empty (zero row) tables, please use the answer posted by Alex Poole.
You could make a table to hold the counts of each table. Then, set a trigger to run on INSERT for each of the tables you're counting that updates the main table.
You'd also need to include a trigger for DELETE.
Basically I have a view which is having below query as a part of view .
SELECT site_id ch_site_id
FROM bfg_router_pi_details
WHERE last_modified_date > (SELECT MAX (last_time_stamp)
FROM saa_bfg_feed_ctl)
Now in this query bfg_router_pi_details is a view in another database and saa_bfg_feed_ctl is a table in same database from we are firing this query.
Now when This query is fired like above using inner query in where clause, it takes 2 hours to finish as it is going over the db link. however if i replace the inner query with the actual date value it takes 2 minutes. so I am trying to see if there is any way that I can define a variable in view and assign the last time stamp value to that variable and then replace that variable in the where clause of the query so that execution become s faster. hope you understood the problem in hand.
It's probably because Oracle is choosing the remote site as the driving site for the query. Try this hint to tell it to run the query from the local site:
SELECT /*+DRIVING_SITE(saa_bfg_feed_ctl) */
site_id ch_site_id
FROM bfg_router_pi_details
WHERE last_modified_date > (SELECT MAX (last_time_stamp)
FROM saa_bfg_feed_ctl)
I'm not 100% sure on this off the top of my head - if you post a plan for your query might be able to help further.
there are many ways how to do it but surely all of them require plsql programming. The most simple one is to create a package where a variable with your MAX is assigned before you make the query to the view.
so the view will look like
begin
SELECT MAX (last_time_stamp)
into your_package.your_variable
FROM saa_bfg_feed_ctl
end;
may be you need no_data_found but it's not my cup of tea
SELECT site_id ch_site_id
FROM bfg_router_pi_details
WHERE last_modified_date > your_package.your_variable
You could use user plsql context by the same way so please consider about it the original documentation - http://docs.oracle.com/cd/B28359_01/network.111/b28531/app_context.htm
I have a view that is very slow if you fetch all rows. But if I select a subset (providing an ID in the where clause) the performance is very good. I cannot hardcode the ID so I create a sub select to get the ID from another table. The sub select only returns one ID. Now the performance is very slow and it seems like Oracle is evaluating the whole view before using the where clause. Can I somehow help Oracle so SQL 2 and 3 have the same performance? I’m using Oracle 10g
1 slow
select * from ci.my_slow_view
2 fast
select * from ci.my_slow_view where id = 1;
3 slow
select * from ci.my_slow_view where id in (select id from active_ids)
How about
select * from ci.my_slow_view where id = (select id from active_ids)
Replacing the "in" with an "=" will tell Oracle that you expect the "select id from active_ids" to return only a single row.
This is the expected behavior...
3 is slow because Oracle will perform a "full table scan", which means that your indexes are not helping there (your where clause does not contain any constant or range and is unbounded, which implies that whatever index you use, all the rows are potentially candidates for the join condition.
Possible improvment:
First, check that the indexes are ok on your join/pk columns (id in my_slow_view and active_ids). This is necessary for the second step:
Second, generate table statistics for your table and views, that will make the Oracle cache memory optimizer kicks in.
(It should work because it is assumed that your active_ids table is small enough to be fully in memory.)
Second approach:
Write a stored procedure in PL/SQL where your id is an in parameter and rewrite your SQL so that it is used a bound parameter.
That should give you the flexibility you need (no hard coded ids), and the speed of the fastest query.
I cannot hardcode the ID so I create a
sub select to get the ID from another
table. The sub select only returns one
ID.
Most likely, gathering statistics on the small table (while it contains a single row) will help, since that should help Oracle realize that it is small and encourage it to use the index on ID.
However, it sounds like this is really not the right solution to your original problem. Generally, when one wants to perform a query repeatedly with a different lookup value, the best way is to use a bind variable. The basic method of doing this in SQLPlus would be:
SQL> variable id number
SQL> exec :id := 1;
SQL> select * from ci.my_slow_view where id = :id ;
SQL> exec :id := 2;
SQL> select * from ci.my_slow_view where id = :id ;
The details of implementing this depend on the environment you are developing in.
Or:
select * from ci.my_slow_view, active_ids
where my_slow_view.id = active_ids.id;