Related
I will try to present my problem as simplified as possible.
Assume that we have 3 tables in Oracle 11g.
Persons (person_id, name, surname, status, etc )
Actions (action_id, person_id, action_value, action_date, calculated_flag)
Calculations (calculation_id, person_id,computed_value,computed_date)
What I want is for each person that meets certain criteria (let's say status=3)
I should get the sum of action_values from the Actions table where calculated_flag=0. (something like this select sum(action_value) from Actions where calculated_flag=0 and person_id=current_id).
Then I shall use that sum in a some kind of formula and update the Calculations table for that specific person_id.
update Calculations set computed_value=newvalue, computed_date=sysdate
where person_id=current_id
After that calculated_flag for participated rows will be set to 1.
update Actions set calculated_flag=1
where calculated_flag=0 and person_id=current_id
Now this can be easily done sequentially, by creating a cursor that will run through Persons table and then execute each action needed for the specific person.
(I don't provide the code for the sequential solution as the above is just an example that resembles my real-world setup.)
The problem is that we are talking about quite big amount of data and sequential approach seems like a waste of computational time.
It seems to me that this task could be performed in parallel for number of person_ids.
So the question is:
Can this kind of task be performed using parallelization in PL/SQL?
What would the solution look like? That is, what special packages (e.g. DBMS_PARALLEL_EXECUTE), keywords (e.g. bulk collect), methods should be used and in what manner?
Also, should I have any concerns about partial failure of parallel updates?
Note that I am not quite familiar with parallel programming with PL/SQL.
Thanks.
Edit 1.
Here my pseudo code for my sequential solution
procedure sequential_solution is
cursor persons_of_interest is
select person_id from persons
where status = 3;
tempvalue number;
newvalue number;
begin
for person in persons_of_interest
loop
begin
savepoint personsp;
--step 1
select sum(action_value) into tempvalue
from actions
where calculated_flag = 0
and person_id = person.person_id;
newvalue := dosomemorecalculations(tempvalue);
--step 2
update calculations set computed_value = newvalue, computed_date = sysdate
where person_id = person.person_id;
--step 3
update actions set calculated_flag = 1;
where calculated_flag = 0 and person_id = person.person_id;
--step 4 (didn't mention this step before - sorry)
insert into actions
( person_id, action_value, action_date, calculated_flag )
values
( person.person_id, 100, sysdate, 0 );
exception
when others then
rollback to personsp;
-- this call is defined with pragma AUTONOMOUS_TRANSACTION:
log_failure(person_id);
end;
end loop;
end;
Now, how would I speed up the above either with forall and bulk colletct or with parallel programming Under the following constrains:
proper memory management (taking into consideration large amount of data)
For a single person if one part of the step sequence fails - all steps should be rolled back and the failure logged.
I can propose the following. Let's say you have 1 000 000 rows in persons table, and you want to process 10 000 persons per iteration. So you can do it in this way:
declare
id_from persons.person_id%type;
id_to persons.person_id%type;
calc_date date := sysdate;
begin
for i in 1 .. 100 loop
id_from := (i - 1) * 10000;
id_to := i * 10000;
-- Updating Calculations table, errors are logged into err$_calculations table
merge into Calculations c
using (select p.person_id, sum(action_value) newvalue
from Actions a join persons p on p.person_id = a.person_id
where a.calculated_flag = 0
and p.status = 3
and p.person_id between id_from and id_to
group by p.person_id) s
on (s.person_id = c.person_id)
when matched then update
set c.computed_value = s.newvalue,
c.computed_date = calc_date
log errors into err$_calculations reject limit unlimited;
-- updating actions table only for those person_id which had no errors:
merge into actions a
using (select distinct p.person_id
from persons p join Calculations c on p.person_id = c.person_id
where c.computed_date = calc_date
and p.person_id between id_from and id_to)
on (c.person_id = p.person_id)
when matched then update
set a.calculated_flag = 1;
-- inserting list of persons for who calculations were successful
insert into actions (person_id, action_value, action_date, calculated_flag)
select distinct p.person_id, 100, calc_date, 0
from persons p join Calculations c on p.person_id = c.person_id
where c.computed_date = calc_date
and p.person_id between id_from and id_to;
commit;
end loop;
end;
How it works:
You split the data in persons table into chunks about 10000 rows (depends on gaps in numbers of ID's, max value of i * 10000 should be knowingly more than maximal person_id)
You make a calculation in the MERGE statement and update the Calculations table
LOG ERRORS clause prevents exceptions. If an error occurs, the row with the error will not be updated, but it will be inserted into a table for errors logging. The execution will not be interrupted. To create this table, execute:
begin
DBMS_ERRLOG.CREATE_ERROR_LOG('CALCULATIONS');
end;
The table err$_calculations will be created. More information about DBMS_ERRLOG package see in the documentation.
The second MERGE statement sets calculated_flag = 1 only for rows, where no errors occured. INSERT statement inserts the these rows into actions table. These rows could be found just with the select from Calculations table.
Also, I added variables id_from and id_to to calculate ID's range to update, and the variable calc_date to make sure that all rows updated in first MERGE statement could be found later by date.
MERGE INTO ////////1 GFO
USING
(SELECT *
FROM
(SELECT facto/////rid,
p-Id,
PRE/////EDATE,
RU//MODE,
cre///date,
ROW_NUMBER() OVER (PARTITION BY facto/////id ORDER BY cre///te DESC) col
FROM ///////////2
) x
WHERE x.col = 1) UFD
ON (GFO.FACTO-/////RID=UFD.FACTO////RID)
WHEN MATCHED THEN UPDATE
SET
GFO.PRE////DATE=UFD.PRE//////DATE
WHERE UFD.CRE/////DATE IS NOT NULL
AND UFD.RU//MODE= 'S'
AND GFO.P////ID=:2
hi every1, my above merge statement is taking too long , it has to run 40 times on table 1 using table2 each having 4millions plus records, for 40 different p--id, please suggest more efficient way as currently its taking 40+ minutes.
its updating only one colummn using a column from table2.t
i am unable to execute the query, its returning
Error: cannot fetch last explain plan from PLAN_TABLE
EXPLAIN PLAN IMAGE
HERE IS THE SCREENSHOT OF EXPLAIN PLAN
cost
The shown plan seems to by OK, the observed problem stems from the LOOP over P_ID that do not scale.
I assume you performs something like this (strongly simplified) - assuming the P_ID to be processed are in table TAB_PID
begin
for cur in (select p_id from tab_pid) loop
merge INTO tab1 USING tab2 ON (tab1.r_id = tab2.r_id)
WHEN MATCHED THEN
UPDATE SET tab1.col1=tab2.col1 WHERE p_id = cur.p_id;
end loop;
end;
/
HASH JOIN on large tables (in NO PARALLEL mode) with elapsed time 60 seconds is not a catastrophic result. But looping 40 times makes your 40 minutes.
So I'd sugesst to try to integrate the loop in the MERGE statement, without knowing details something like this (mayby you'll need also ajdust the MERGE JOIN condition).
merge INTO tab1 USING tab2 ON (tab1.r_id = tab2.r_id)
WHEN MATCHED THEN
UPDATE SET tab1.col1=tab2.col1
WHERE p_id in (select p_id from tab_pid);
I have 2 delete statements that are taking a long time to complete. There are several indexes on the columns in where clause.
What is a duplicate?
If 2 or more records have same values in columns id,cid,type,trefid,ordrefid,amount and paydt then there are duplicates.
The DELETEs delete about 1 million record.
Can they be re-written in any way to make it quicker.
DELETE FROM TABLE1 A WHERE loaddt < (
SELECT max(loaddt) FROM TABLE1 B
WHERE
a.id=b.id and
a.cid=b.cid and
NVL(a.type,'-99999') = NVL(b.type,'-99999') and
NVL(a.trefid,'-99999')=NVL(b.trefid,'-99999') and
NVL(a.ordrefid,'-99999')= NVL(b.ordrefid,'-99999') and
NVL(a.amount,'-99999')=NVL(b.amount,'-99999') and
NVL(a.paydt,TO_DATE('9999-12-31','YYYY-MM-DD'))=NVL(b.paydt,TO_DATE('9999-12-31','YYYY-MM-DD'))
);
COMMIT;
DELETE FROM TABLE1 a where rowid > (
Select min(rowid) from TABLE1 b
WHERE
a.id=b.id and
a.cid=b.cid and
NVL(a.type,'-99999') = NVL(b.type,'-99999') and
NVL(a.trefid,'-99999')=NVL(b.trefid,'-99999') and
NVL(a.ordrefid,'-99999')= NVL(b.ordrefid,'-99999') and
NVL(a.amount,'-99999')=NVL(b.amount,'-99999') and
NVL(a.paydt,TO_DATE('9999-12-31','YYYY-MM-DD'))=NVL(b.paydt,TO_DATE('9999-12-31','YYYY-MM-DD'))
);
commit;
Explain Plan:
DELETE TABLE1
HASH JOIN 1296491
Access Predicates
AND
A.ID=ITEM_1
A.CID=ITEM_2
ITEM_3=NVL(TYPE,'-99999')
ITEM_4=NVL(TREFID,'-99999')
ITEM_5=NVL(ORDREFID,'-99999')
ITEM_6=NVL(AMOUNT,(-99999))
ITEM_7=NVL(PAYDT,TO_DATE(' 9999-12-31 00:00:00', 'syyyy-mm-dd hh24:mi:ss'))
Filter Predicates
LOADDT<MAX(LOADDT)
TABLE ACCESS TABLE1 FULL 267904
VIEW VW_SQ_1 690385
SORT GROUP BY 690385
TABLE ACCESS TABLE1 FULL 267904
How large is the table? If count of deleted rows is up to 12% then you may think about index.
Could you somehow partition your table - like week by week and then scan only actual week?
Maybe this could be more effecient. When you're using aggregate function, then oracle must walk through all relevant rows (in your case fullscan), but when you use exists it stops when the first occurence is found. (and of course the query would be much faster, when there was one function-based(because of NVL) index on all columns in where clause)
DELETE FROM TABLE1 A
WHERE exists (
SELECT 1
FROM TABLE1 B
WHERE
A.loaddt != b.loaddt
a.id=b.id and
a.cid=b.cid and
NVL(a.type,'-99999') = NVL(b.type,'-99999') and
NVL(a.trefid,'-99999')=NVL(b.trefid,'-99999') and
NVL(a.ordrefid,'-99999')= NVL(b.ordrefid,'-99999') and
NVL(a.amount,'-99999')=NVL(b.amount,'-99999') and
NVL(a.paydt,TO_DATE('9999-12-31','YYYY-MM-DD'))=NVL(b.paydt,TO_DATE('9999-12-31','YYYY-MM-DD'))
);
Although some may disagree, I am a proponent of running large, long running deletes procedurally. In my view it is much easier to control and track progress (and your DBA will like you better ;-) Also, not sure why you need to join table1 to itself to identify duplicates (and I'd be curious if you ever run into snapshot too old issues with your current approach). You also shouldn't need multiple delete statements, all duplicates should be handled in one process. Finally, you should check WHY you're constantly re-introducing duplicates each week, and perhaps change the load process (maybe doing a merge/upsert rather than all inserts).
That said, you might try something like:
-- first create mat view to find all duplicates
create materialized view my_dups_mv
tablespace my_tablespace
build immediate
refresh complete on demand
as
select id,cid,type,trefid,ordrefid,amount,paydt, count(1) as cnt
from table1
group by id,cid,type,trefid,ordrefid,amount,paydt
having count(1) > 1;
-- dedup data (or put into procedure and schedule along with mat view refresh above)
declare
-- make sure my_dups_mv is refreshed first
cursor dup_cur is
select * from my_dups_mv;
type duprec_t is record(row_id rowid);
duprec duprec_t;
type duptab_t is table of duprec_t index by pls_integer;
duptab duptab_t;
l_ctr pls_integer := 0;
l_dupcnt pls_integer := 0;
begin
for rec in dup_cur
loop
l_ctr := l_ctr + 1;
-- assuming needed indexes exist
select rowid
bulk collect into duptab
from table1
where id = rec.id
and cid = rec.cid
and type = rec.type
and trefid = rec.trefid
and ordrefid = rec.ordrefid
and amount = rec.amount
and paydt = rec.paydt
-- order by whatever makes sense to make the "keeper" float to top
order by loaddt desc
;
for i in 2 .. duptab.count
loop
l_dupcnt := l_dupcnt + 1;
delete from table1 where rowid = duptab(i).row_id;
end loop;
if (mod(l_ctr, 10000) = 0) then
-- log to log table here (calling autonomous procedure you'll need to implement)
insert_logtable('Table1 deletes', 'Commit reached, deleted ' || l_dupcnt || ' rows');
commit;
end if;
end loop;
commit;
end;
Check your log table for progress status.
1. Parallel
alter session enable parallel dml;
DELETE /*+ PARALLEL */ FROM TABLE1 A WHERE loaddt < (
...
Assuming you have Enterprise Edition, a sane server configuration, and you are on 11g. If you're not on 11g, the parallel syntax is slightly different.
2. Reduce memory requirements
The plan shows a hash join, which is probably a good thing. But without any useful filters, Oracle has to hash the entire table. (Tbone's query, that only use a GROUP BY, looks nicer and may run faster. But it will also probably run into the same problem trying to sort or hash the entire table.)
If the hash can't fit in memory it must be written to disk, which can be very slow. Since you run this query every week, only one of the tables needs to look at all the rows. Depending on exactly when it runs, you can add something like this to the end of the query: ) where b.loaddt >= sysdate - 14. This may significantly reduce the amount of writing to temporary tablespace. And it may also reduce read IO if you use some partitioning strategy like jakub.petr suggested.
3. Active Report
If you want to know exactly what your query is doing, run the Active Report:
select dbms_sqltune.report_sql_monitor(sql_id => 'YOUR_SQL_ID_HERE', type => 'active')
from dual;
(Save the output to an .html file and open it with a browser.)
I asked a question yesterday which got answers but didnt answer the main point. I wanted to reduce amount of time it took to do a MINUS operation.
Now, I'm thinking about doing MINUS operation in blocks of 5000, appending each iterations results to the cursor and finally returning the cursor.
I have following:
V_CNT NUMBER :=0;
V_INTERVAL NUMBER := 5000;
begin
select count(1) into v_cnt from TABLE_1
while (v_cnt > 0)
loop
open cv_1 for
SELECT A.HEAD,A.EFFECTIVE_DATE,
FROM TABLE_1 A
WHERE A.TYPE_OF_ACTION='6' AND A.EFFECTIVE_DATE >= ADD_MONTHS(SYSDATE,-15)
AND A.ROWNUM <= V_INTERVAL
MINUS
SELECT B.head,B.EFFECTIVE_DATE,
FROM TABLE_2 B
AND B.ROWNUM <= V_INTERVAL
V_CNT := V_CNT - V_INTERVAL;
END LOOP;
end;
However, as you see...in each iteration the cursor is overwritten. How can I change the code so that in each iteration it appends to cv_1 cursor rather than overwriting?
You haven't stated the requirement clearly.
So , i am assuming , you want to do a MINUS on two tables, A and B.
i.e you want to find tuples in A that are not in B.
Assuming this , the logic that you have written is not completely correct, as you are doing a MINUS on corresponding (5000-length) batches of A and B.
Eg: Your logic will return a tuple in the 4000th row in table A, that is present in say the 6000th row of table B.
I suggest you use left-outer join to accomplish your need. (Same as Peter Lang's post).
That should suffice for your performance requirements too, I think.
That's not how cursors work, you would have to store the values in some sort of collection.
Your current query gets you 5000 random rows from Table_1 and removes rows that also exist in 5000 random rows selected from Table_2.
Have you tried doing it without the MINUS?
As I understand the query, it should produce the same as this one:
Select a.head, a.effective_date,
From table_1 a
Left Join table_2 b On (b.head = a.head And b.effective_date = a.effective_date )
Where a.type_of_action='6' And a.effective_date >= ADD_MONTHS(SYSDATE,-15)
And b.head Is Null;
Having a compound index on TABLE_1 (type_of_action, head, effective_date) and on TABLE_2 (head, effective_date) should help you with performance.
This question already has answers here:
SQL IN Clause 1000 item limit
(5 answers)
Closed 8 years ago.
Is there any way to get around the Oracle 10g limitation of 1000 items in a static IN clause? I have a comma delimited list of many of IDs that I want to use in an IN clause, Sometimes this list can exceed 1000 items, at which point Oracle throws an error. The query is similar to this...
select * from table1 where ID in (1,2,3,4,...,1001,1002,...)
Put the values in a temporary table and then do a select where id in (select id from temptable)
select column_X, ... from my_table
where ('magic', column_X ) in (
('magic', 1),
('magic', 2),
('magic', 3),
('magic', 4),
...
('magic', 99999)
) ...
I am almost sure you can split values across multiple INs using OR:
select * from table1 where ID in (1,2,3,4,...,1000) or
ID in (1001,1002,...,2000)
You may try to use the following form:
select * from table1 where ID in (1,2,3,4,...,1000)
union all
select * from table1 where ID in (1001,1002,...)
Where do you get the list of ids from in the first place? Since they are IDs in your database, did they come from some previous query?
When I have seen this in the past it has been because:-
a reference table is missing and the correct way would be to add the new table, put an attribute on that table and join to it
a list of ids is extracted from the database, and then used in a subsequent SQL statement (perhaps later or on another server or whatever). In this case, the answer is to never extract it from the database. Either store in a temporary table or just write one query.
I think there may be better ways to rework this code that just getting this SQL statement to work. If you provide more details you might get some ideas.
Use ...from table(... :
create or replace type numbertype
as object
(nr number(20,10) )
/
create or replace type number_table
as table of numbertype
/
create or replace procedure tableselect
( p_numbers in number_table
, p_ref_result out sys_refcursor)
is
begin
open p_ref_result for
select *
from employees , (select /*+ cardinality(tab 10) */ tab.nr from table(p_numbers) tab) tbnrs
where id = tbnrs.nr;
end;
/
This is one of the rare cases where you need a hint, else Oracle will not use the index on column id. One of the advantages of this approach is that Oracle doesn't need to hard parse the query again and again. Using a temporary table is most of the times slower.
edit 1 simplified the procedure (thanks to jimmyorr) + example
create or replace procedure tableselect
( p_numbers in number_table
, p_ref_result out sys_refcursor)
is
begin
open p_ref_result for
select /*+ cardinality(tab 10) */ emp.*
from employees emp
, table(p_numbers) tab
where tab.nr = id;
end;
/
Example:
set serveroutput on
create table employees ( id number(10),name varchar2(100));
insert into employees values (3,'Raymond');
insert into employees values (4,'Hans');
commit;
declare
l_number number_table := number_table();
l_sys_refcursor sys_refcursor;
l_employee employees%rowtype;
begin
l_number.extend;
l_number(1) := numbertype(3);
l_number.extend;
l_number(2) := numbertype(4);
tableselect(l_number, l_sys_refcursor);
loop
fetch l_sys_refcursor into l_employee;
exit when l_sys_refcursor%notfound;
dbms_output.put_line(l_employee.name);
end loop;
close l_sys_refcursor;
end;
/
This will output:
Raymond
Hans
I wound up here looking for a solution as well.
Depending on the high-end number of items you need to query against, and assuming your items are unique, you could split your query into batches queries of 1000 items, and combine the results on your end instead (pseudocode here):
//remove dupes
items = items.RemoveDuplicates();
//how to break the items into 1000 item batches
batches = new batch list;
batch = new batch;
for (int i = 0; i < items.Count; i++)
{
if (batch.Count == 1000)
{
batches.Add(batch);
batch.Clear()
}
batch.Add(items[i]);
if (i == items.Count - 1)
{
//add the final batch (it has < 1000 items).
batches.Add(batch);
}
}
// now go query the db for each batch
results = new results;
foreach(batch in batches)
{
results.Add(query(batch));
}
This may be a good trade-off in the scenario where you don't typically have over 1000 items - as having over 1000 items would be your "high end" edge-case scenario. For example, in the event that you have 1500 items, two queries of (1000, 500) wouldn't be so bad. This also assumes that each query isn't particularly expensive in of its own right.
This wouldn't be appropriate if your typical number of expected items got to be much larger - say, in the 100000 range - requiring 100 queries. If so, then you should probably look more seriously into using the global temporary tables solution provided above as the most "correct" solution. Furthermore, if your items are not unique, you would need to resolve duplicate results in your batches as well.
Yes, very weird situation for oracle.
if you specify 2000 ids inside the IN clause, it will fail.
this fails:
select ...
where id in (1,2,....2000)
but if you simply put the 2000 ids in another table (temp table for example), it will works
below query:
select ...
where id in (select userId
from temptable_with_2000_ids )
what you can do, actually could split the records into a lot of 1000 records and execute them group by group.
Here is some Perl code that tries to work around the limit by creating an inline view and then selecting from it. The statement text is compressed by using rows of twelve items each instead of selecting each item from DUAL individually, then uncompressed by unioning together all columns. UNION or UNION ALL in decompression should make no difference here as it all goes inside an IN which will impose uniqueness before joining against it anyway, but in the compression, UNION ALL is used to prevent a lot of unnecessary comparing. As the data I'm filtering on are all whole numbers, quoting is not an issue.
#
# generate the innards of an IN expression with more than a thousand items
#
use English '-no_match_vars';
sub big_IN_list{
#_ < 13 and return join ', ',#_;
my $padding_required = (12 - (#_ % 12)) % 12;
# get first dozen and make length of #_ an even multiple of 12
my ($a,$b,$c,$d,$e,$f,$g,$h,$i,$j,$k,$l) = splice #_,0,12, ( ('NULL') x $padding_required );
my #dozens;
local $LIST_SEPARATOR = ', '; # how to join elements within each dozen
while(#_){
push #dozens, "SELECT #{[ splice #_,0,12 ]} FROM DUAL"
};
$LIST_SEPARATOR = "\n union all\n "; # how to join #dozens
return <<"EXP";
WITH t AS (
select $a A, $b B, $c C, $d D, $e E, $f F, $g G, $h H, $i I, $j J, $k K, $l L FROM DUAL
union all
#dozens
)
select A from t union select B from t union select C from t union
select D from t union select E from t union select F from t union
select G from t union select H from t union select I from t union
select J from t union select K from t union select L from t
EXP
}
One would use that like so:
my $bases_list_expr = big_IN_list(list_your_bases());
$dbh->do(<<"UPDATE");
update bases_table set belong_to = 'us'
where id in ($bases_list_expr)
UPDATE
Instead of using IN clause, can you try using JOIN with the other table, which is fetching the id. that way we don't need to worry about limit. just a thought from my side.
Instead of SELECT * FROM table1 WHERE ID IN (1,2,3,4,...,1000);
Use this :
SELECT * FROM table1 WHERE ID IN (SELECT rownum AS ID FROM dual connect BY level <= 1000);
*Note that you need to be sure the ID does not refer any other foreign IDS if this is a dependency. To ensure only existing ids are available then :
SELECT * FROM table1 WHERE ID IN (SELECT distinct(ID) FROM tablewhereidsareavailable);
Cheers