Parallel Execution questions

Hi everyone, i have few questions about Parallel Execution.
1- Query Coordinator to scan the table, split up the table to the processes. But, each process take different number of rows. For example, in oracle 10g database architecture, thomas kyte gave an example:
"SELECT COUNT(*) FROM BIG_TABLE"
It is distributed to 4 processes. One process take 1000 rows, another process 1200 etc. Why this is not equal? How the number of rows is determined by the QC?
2- If we do not determine the degree of parallelism ( /*+ PARALLEL(B) */ instead of /*+ PARALLEL(B,16) */ ) how the cbo determine the degree of parallelism?
3- in the explain plan i see lots of thing i do not know. For example :
:TQ10000
what is that?
-> TQ, IN-OUT, PQ_Disturb columns?
thanks for responses.

Hi
Why this is not equal? How the number of rows is determined by the QC?The number of rows is not used to split the table. Data is split in granules and distributed to the slave processed based on two methods:
- partition granules: a whole partition is given to a specific slave process
- block range granules: range of blocks are given to each slave process
So, depending on the size of the granules, one slave process might process much more data than another.
If we do not determine the degree of parallelism how the cbo determine
the degree of parallelism?It depends on the configuration. In fact, there is a default at session level, at system level and at table level. Depending which ones are set, a DOP is chosen. In addition, the DOP might be decreased at runtime (before the execution starts, not dynamically during it).
For example : :TQ10000 what is that?Producers send data to consumers through so called "table queues" (TQ). The number, in your case 10000, is just a identifier generated by the database engine.
TQTable queue, e.g.: TQ10000 -> Q1,00
IN-OUTThis is the relationship between parallel operations. E.g. P->P means that a parallel operation sends data to another parallel operation.
PQ_Disturb columns?This is the method used to distribute rows. E.g. RANGE means that the producers send specific ranges of rows to different consumers.
For a more detailed discussion about these topics have a look to the documentation:
http://download.oracle.com/docs/cd/B19306_01/server.102/b14223/usingpe.htm#i1009828
HTH
Chris Antognini
Author of Troubleshooting Oracle Performance, Apress 2008 (http://top.antognini.ch)

Similar Messages

11g Parallel Execution on AIX 6 - SMT Enabled or Disabled?

Greetings,
I've had no luck searching for an answer to this question and I'm hoping someone can answer it:
Can Oracle 11g Parallel Execution spread the "granules" of paralllism across the threads (logical cpus) in an AIX SMT enabled environment, or should SMT be disabled and the "granules" be spread scross the processors (virtual cpus)? The application is a data warehouse in a non-RAC configuration using a p570 server. From what I've read, the server must be SMP for Oracle parallel execution capabilities to be maximized, but I think all AIX servers are SMP (not sure if the server needs to be ordered as an SMP server). I'm mostly concerned with the data load processing at this point, and not so concerned for the query right now. I believe AIX 6.1 can enable/disable SMT dynamically. So would it make sense to disable during data loads, and enable for DSS query?
Hope the question makes sense. Thanks for any help in advance!

SMT will determine Oracle's cpu_count parameter.
However this is a static parameter.
So it won't work, and it might be even dangerous to change it on the fly.
Sybrand Bakker
Senior Oracle DBA
Experts: those who did read documentation.

Oracle 10g and parallel query question

Hi Oracle on SAP Gurus!
We are currently thinking of activating parallel query for certain segments (large application tables and indexes). We searched in SAPNet and SDN and have also studied the SAP Note 651060. But we did not find a complete answer to the following question which is very important for us:
Which kinds of queries (despite from full table scan and index scan in partitioned indexes) support parallel queries and which ones do not support parallel queries?
This is important for us to find out whether we have candidates for parallel queries or not.
Thanx for any hint!
Regards,
Volker

But why do you not propose to use parallel query in OLTP systems?
If the queries are accessed very frequently you will just run out of cpu and io ressources. OLTP systems are (historical) typically multi user systems. You can off course use PQ for 'single user' activities, like index rebuilds, some batchjobs, but you shouldn't do for frequent user queries.
If you have time look at this interesting Article [Suck It Dry - Tuning Parallel Execution|http://doug.burns.tripod.com/px.html]
It is quite old, and you don't have to read all tech details, but i recommend having a look at the conclusions at the end.
May it make sense to use partitioning of these tables in conjunction with parallel query?
I know some guys, who do partitioning on OLTP systems, even SAP systems. But they don't use PQ then. The use partitioning to work on a smaller set of data. In your case the range scans, would need to scan only one partition, saving buffer cache and effectively speeding up execution. So you don't need PQ to scan all partitions at all, this would be a typical OLAP approach.
Best regards
Michael

Parallel execution of multiple scripts

Hi All,
I understand that by using parallel hint we can achieve parallel execution of the queries(insert/update/delete).
For my today's question I would like know if Parallelism can be achieved for the following scenario.
I have a script with an insert-select statement and multiple merge statements and few update statements all on the same table.
I have to run this script 12 times on the same table, once for each month of the year.
Currently we run for Jan (where record_date = '201001') commit it, and then run for Feb, and so on.
Can all the 12 month be run in parallel. One way I can think of is create 12 different scripts and kick them of by opening 12 different SQL Plus sessions (all on the same table).
Is there a better way of doing this ?
Note: each month data will not effect the other other month data.
Regards,
Aj

Creating 12 different scripts would be a sub-optimal solution, and a maintenance nightmare. Creating 1 parametrized script and call that 12 times with different parameters would be slightly better.
Creating 1 stored procedure with parameters, and run that procedure in the scheduler 12 times would still be better, as everything runs at the server side only.
Your description is deliberately vague, so the possibility of deadlock can not be excluded.
Sybrand Bakker
Senior Oracle DBA

Parallel execution

Hello!
One of my views includes
"decode" in which one of expressions is "select from ... " ,
all indexes and tables have degree of parallelism equal 1,
and only sometimes I see parallel execution of guery.
My question is
whether can my decode's become the reason of parallel execution?
Thanks and regards,
Pavel

Hi Pavel,
Pavel wrote:
Hello!
One of my views includes
"decode" in which one of expressions is "select from ... " ,
all indexes and tables have degree of parallelism equal 1,
and only sometimes I see parallel execution of guery.
My question is
whether can my decode's become the reason of parallel execution?No, decision to run a statement in parallel is taken based on many considerations, e.g. degree of parallelism of the objects involved, session settings (parallel query/parallel DML), cost etc. but use of decode function isn't one of them.
Best regards,
Nikolay

Query Cubes with Parallel executions?

I have the following problem:
As I mentioned in my first question here: Combined Cube of smaller cubes?
We joined four cube views to form a fact table for a bigger cube. Actually we don't form a bigger view based on the cube views' now, but use an etl to insert into a fact table the results by months. Let's say our cubes are DDS_CUBE, KD_CUBE, GFO_CUBE and ZL_CUBE.
We use DDS_CUBE as our main and do a left join on the others to insert rows in fact table for AOR_CUBE. There are around 238 000 rows per month in DDS_CUBE and DDS_CUBE have data for 12 months. The other cubes have average the same rows of data.
And when I ran the query with all the rows after 3 hours it's still running and I think it shouldn't be.
*When I limit the rows with rownum < 1000 it executes for under 5 minutes.
Is there a way to optimize this, can I query the cubes with parallel executions not from the views but from the table they are kept?
Edit: Forgot to mention that DDS_CUBE is partitioned by month and have a total of 84 partitions, but only 12 are loaded.
Message was edited by: valeksandrova
Also is there a way to import OLAP cubes from AWM to OWB?

Update:
I wrote a procedure with bulk collect. It turns out that the DDS_CUBE is causing problems. The query selects from it around 238k rows which are the total amount of rows in the cube for the selected month and then it just stops, it doesn't proceed the rest of the query to select from the other cubes.
Also, lets say my select statement is like this:
SELECT <cols> FROM DDS_CUBE
LEFT JOIN GFO_CUBE on <join>
LEFT JOIN KD_CUBE on <join>
LEFT JOIN ZL_CUBE on <join>
WHERE <condition that select only the members of the lowest dimension levels and for only one month>
And I'm wondering why does the execution plan does right join on the GFO, KD and ZL cubes and not a LEFT join based on the DDS_CUBE.
It goes selecting all(all months, all levels, everything) the rows of the other cubes (11M from ZL and 238k from DDS, and there it stops, doesn't proceed to KD or GFO cube). Why isn't just selecting first the needed 238k from DDS_CUBE and then filter only the needed ones from the other cubes?

ODISqlUnload - Parallel Execution

Hello, I have developed a KM that uses ODISqlUnload to extract data from an Oracle source. I have noticed that when I attempt to run 2 jobs using the same KM, the first job fails when the second job is executed. For example:
Session 1 : Executed at 12:20
Session 2 : Executed at 12:30
When Session 2 executes the ODISqlUnload command, Session 1 will fail but Session 2 will continue.
My question is, is there a parameter to instruct ODISqlUnload to enable parallel execution? Or is it a limitation of the tool?
Thanks in advance!

smercurio_fc wrote:
Well, there you go then. You can only have 1 "read" task for one device active at a time. Your "78652-manual_EDIT" is set up in parallel, but each task can only execute once the currently active "read" is done. Since the "7862SUBVI_manualmode" VI is dependent on data from the "78652-manual_EDIT" VI, it can only execute once the data has been read. You can mitigate this somewhat so that instead of 6 separate reads you perform 1 read to get a 2D array of booleans and then index out each one to your 6 instances of "7862SUBVI_manualmode". The "7862SUBVI_manualmode" VI, however, contains additional DAQ tasks so you may get some blocking in there as well.
What I am trying to say is, you said I should create an array and perform the reads. I had done that initally. But the subvi contains DAQ tasks and using the value after indexing arrays is leading to errors.
Also the P3/0-3.5 ets are port nos and lines. So i don't have lines of a function in a straight order together. The indexing also makes DAQ writes difficult to implement.

11gR2 query parallel execution

I need to retrieve info from a very big table many times with different where conditions and compared execution plan for the queries using composite index on (group_id, start,end).
SELECT /*+ index(a idx_loc) */ a.id , a.group_id
from location a
where a.group_id=10
and a.start>=30000
and a.end<=60000;
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | Pstart| Pstop |
| 0 | SELECT STATEMENT | | 578 | 25432 | 191K (1)| 00:38:14 | | |
| 1 | PARTITION LIST SINGLE | | 578 | 25432 | 191K (1)| 00:38:14 | KEY | KEY |
| 2 | PARTITION LIST SINGLE | | 578 | 25432 | 191K (1)| 00:38:14 | 6 | 6 |
| 3 | TABLE ACCESS BY LOCAL INDEX ROWID| LOCATION | 578 | 25432 | 191K (1)| 00:38:14 | 6 | 6 |
|* 4 | INDEX SKIP SCAN | IDX_LOC | 592 | | 190K (1)| 00:38:07 | 6 | 6 |
SELECT /*+ parallel */ a.id , a.group_id
from location a
where a.group_id=10
and a.start>=30000
and a.end<=60000;
| Id | Operation | Name | Rows | Bytes | Cost (%CPU)| Time | Pstart| Pstop | TQ |IN-OUT| PQ Distrib |
| 0 | SELECT STATEMENT | | 578 | 25432 | 16333 (1)| 00:00:33 | | | | | |
| 1 | PX COORDINATOR | | | | | | | | | | |
| 2 | PX SEND QC (RANDOM)| :TQ10000 | 578 | 25432 | 16333 (1)| 00:00:33 | | | Q1,00 | P->S | QC (RAND) |
| 3 | PX BLOCK ITERATOR | | 578 | 25432 | 16333 (1)| 00:00:33 | 6 | 6 | Q1,00 | PCWC | |
|* 4 | TABLE ACCESS FULL| LOCATION | 578 | 25432 | 16333 (1)| 00:00:33 | 6 | 6 | Q1,00 | PCWP | |
My questions are:
1) The execution plan shows the query using parallel hint has much less cpu cost and time compared with the one with index hint. So it seems it is better to use the parallel execution. is there any downside to use parallel hint that I should consider?
2) I can also modify the application to submit multiple queries (using index hint) in parallel (which opens multiple sessions). Is the parallel hint in the query superior to parallel queries in application?

user7435395 wrote:
I need to retrieve info from a very big table many times with different where conditions and compared execution plan for the queries using composite index on (group_id, start,end). BIG is a very qualitative term. So you need to be more specific. And when ever you ask for a SQL Tuning request please provide the following details mentioned in {thread:id=863295}
SELECT /*+ index(a idx_loc) */ a.id , a.group_id
from location a
where a.group_id=10
and a.start>;=30000
and a.end;=60000;
| Id | Operation                           | Name                      | Rows | Bytes | Cost (%CPU)| Time     | Pstart| Pstop |
|   0 | SELECT STATEMENT                    |                           |   578 | 25432 |   191K (1)| 00:38:14 |       |       |
|   1 | PARTITION LIST SINGLE              |                           |   578 | 25432 |   191K (1)| 00:38:14 |   KEY |   KEY |
|   2 |   PARTITION LIST SINGLE             |                           |   578 | 25432 |   191K (1)| 00:38:14 |     6 |     6 |
|   3 |    TABLE ACCESS BY LOCAL INDEX ROWID| LOCATION |   578 | 25432 |   191K (1)| 00:38:14 |     6 |     6 |
|* 4 |     INDEX SKIP SCAN                 | IDX_LOC           |   592 |       |   190K (1)| 00:38:07 |     6 |     6 |
--------------------------------------------------------------------------------------------------------------------------------- As you have forced oracle to use IDX_LOC. it is going for a [url http://docs.oracle.com/cd/B19306_01/server.102/b14211/optimops.htm#PFGRF10105]INDEX SKIP SCAN. And if the cardinality of the subindex column is low then oracle may be doing more work here. I think that is what exactly happening in your case. A simple full table scan could be more efficient.
SELECT /*+ parallel */ a.id , a.group_id
from location a
where a.group_id=10
and a.start>;=30000
and a.end;=60000;
| Id | Operation            | Name                      | Rows | Bytes | Cost (%CPU)| Time     | Pstart| Pstop |    TQ |IN-OUT| PQ Distrib |
|   0 | SELECT STATEMENT     |                           |   578 | 25432 | 16333   (1)| 00:00:33 |       |       |        |      |            |
|   1 | PX COORDINATOR      |                           |       |       |            |          |       |       |        |      |            |
|   2 |   PX SEND QC (RANDOM)| :TQ10000                  |   578 | 25432 | 16333   (1)| 00:00:33 |       |       | Q1,00 | P->S | QC (RAND) |
|   3 |    PX BLOCK ITERATOR |                           |   578 | 25432 | 16333   (1)| 00:00:33 |     6 |     6 | Q1,00 | PCWC |            |
|* 4 |     TABLE ACCESS FULL| LOCATION |   578 | 25432 | 16333   (1)| 00:00:33 |     6 |     6 | Q1,00 | PCWP |            |
----------------------------------------------------------------------------------------------------------------------------------------------- Again PARALLEL query has its own down side. There is no free lunch. You need to spend more resource. And most of the time in a OLTP system with multi user access its not worthwhile.
So the first thing i would suggest you is to remove all your hint and get an execution plan and post it. And read the above link provided an post all the necessary details.
You need to be first clear about your problem.

Parallel execution and temporary tablespaces

I have a large long running (1 hour) data warehouse query in a materialized view.
If I parallelize it using the parallel hint then I run out of temporary tablespace.
I've tried creating a bunch of temporary tablespaces and putting them into a temp tablespace group but it still runs out of space. Parallel execution seems to use up way more temp tablespace than sequential execution.
I know it is a very general question, but what are the tips for parallelizing a long running query with respect to temporary tablespace management?
I've tried searching on the interwebs but I don't find anything that addresses this particular issue.

And here is the parallel explain plan:
PLAN_TABLE_OUTPUT
Plan hash value: 1293981491
| Id | Operation | Name | Rows | Bytes |TempSpc| Cost (%CPU)| Time | TQ |IN-OUT| PQ Distrib |
| 0 | SELECT STATEMENT | | 81M| 17G| | 19232 (2)| 00:00:01 | | | |
| 1 | TABLE ACCESS BY INDEX ROWID | MART$SC_SCORES | 1 | 13 | | 3 (0)| 00:00:01 | | | |
|* 2 | INDEX UNIQUE SCAN | SSCS_SDCC_FK_I | 1 | | | 2 (0)| 00:00:01 | | | |
| 3 | PX COORDINATOR | | | | | | | | | |
| 4 | PX SEND QC (RANDOM) | :TQ10017 | 81M| 17G| | 19232 (2)| 00:00:01 | Q1,17 | P->S | QC (RAND) |
|* 5 | HASH JOIN RIGHT OUTER BUFFERED | | 81M| 17G| | 19232 (2)| 00:00:01 | Q1,17 | PCWP | |
| 6 | PX RECEIVE | | 41925 | 491K| | 2 (0)| 00:00:01 | Q1,17 | PCWP | |
| 7 | PX SEND BROADCAST | :TQ10014 | 41925 | 491K| | 2 (0)| 00:00:01 | Q1,14 | P->P | BROADCAST |
| 8 | PX BLOCK ITERATOR | | 41925 | 491K| | 2 (0)| 00:00:01 | Q1,14 | PCWC | |
| 9 | INDEX FAST FULL SCAN | I_DWH_ZIP_ZIPCODE_EIDIID | 41925 | 491K| | 2 (0)| 00:00:01 | Q1,14 | PCWP | |
|* 10 | HASH JOIN | | 81M| 16G| | 19218 (2)| 00:00:01 | Q1,17 | PCWP | |
| 11 | JOIN FILTER CREATE | :BF0000 | 6414K| 159M| | 806 (2)| 00:00:01 | Q1,17 | PCWP | |
| 12 | PX RECEIVE | | 6414K| 159M| | 806 (2)| 00:00:01 | Q1,17 | PCWP | |
| 13 | PX SEND HASH | :TQ10015 | 6414K| 159M| | 806 (2)| 00:00:01 | Q1,15 | P->P | HASH |
| 14 | PX BLOCK ITERATOR | | 6414K| 159M| | 806 (2)| 00:00:01 | Q1,15 | PCWC | |
|* 15 | INDEX FAST FULL SCAN | I_DWH_ADDRESS_COMB_ZIP | 6414K| 159M| | 806 (2)| 00:00:01 | Q1,15 | PCWP | |
| 16 | PX RECEIVE | | 80M| 14G| | 18397 (2)| 00:00:01 | Q1,17 | PCWP | |
| 17 | PX SEND HASH | :TQ10016 | 80M| 14G| | 18397 (2)| 00:00:01 | Q1,16 | P->P | HASH |
| 18 | JOIN FILTER USE | :BF0000 | 80M| 14G| | 18397 (2)| 00:00:01 | Q1,16 | PCWP | |
|* 19 | HASH JOIN RIGHT OUTER BUFFERED | | 80M| 14G| | 18397 (2)| 00:00:01 | Q1,16 | PCWP | |
| 20 | PX RECEIVE | | 42M| 409M| | 827 (2)| 00:00:01 | Q1,16 | PCWP | |
| 21 | PX SEND HASH | :TQ10012 | 42M| 409M| | 827 (2)| 00:00:01 | Q1,12 | P->P | HASH |
| 22 | PX BLOCK ITERATOR | | 42M| 409M| | 827 (2)| 00:00:01 | Q1,12 | PCWC | |
| 23 | MAT_VIEW ACCESS FULL | MBI$CMN_ACTION_COST | 42M| 409M| | 827 (2)| 00:00:01 | Q1,12 | PCWP | |
| 24 | PX RECEIVE | | 80M| 14G| | 17549 (1)| 00:00:01 | Q1,16 | PCWP | |
| 25 | PX SEND HASH | :TQ10013 | 80M| 14G| | 17549 (1)| 00:00:01 | Q1,13 | P->P | HASH |
|* 26 | HASH JOIN BUFFERED | | 80M| 14G| | 17549 (1)| 00:00:01 | Q1,13 | PCWP | |
| 27 | PX RECEIVE | | 6312K| 794M| | 7519 (1)| 00:00:01 | Q1,13 | PCWP | |
| 28 | PX SEND HASH | :TQ10010 | 6312K| 794M| | 7519 (1)| 00:00:01 | Q1,10 | P->P | HASH |
|* 29 | HASH JOIN RIGHT OUTER BUFFERED | | 6312K| 794M| | 7519 (1)| 00:00:01 | Q1,10 | PCWP | |
| 30 | VIEW | | 4443K| 80M| | 2125 (2)| 00:00:01 | Q1,10 | PCWP | |
| 31 | HASH GROUP BY | | 4443K| 33M| 158M| 2125 (2)| 00:00:01 | Q1,10 | PCWP | |
| 32 | PX RECEIVE | | 10M| 78M| | 950 (1)| 00:00:01 | Q1,10 | PCWP | |
| 33 | PX SEND HASH | :TQ10007 | 10M| 78M| | 950 (1)| 00:00:01 | Q1,07 | P->P | HASH |
| 34 | PX BLOCK ITERATOR | | 10M| 78M| | 950 (1)| 00:00:01 | Q1,07 | PCWC | |
| 35 | TABLE ACCESS FULL | DWH$PHONE | 10M| 78M| | 950 (1)| 00:00:01 | Q1,07 | PCWP | |
|* 36 | HASH JOIN | | 6312K| 680M| | 5392 (1)| 00:00:01 | Q1,10 | PCWP | |
| 37 | PX RECEIVE | | 6329K| 36M| | 130 (2)| 00:00:01 | Q1,10 | PCWP | |
| 38 | PX SEND HASH | :TQ10008 | 6329K| 36M| | 130 (2)| 00:00:01 | Q1,08 | P->P | HASH |
| 39 | PX BLOCK ITERATOR | | 6329K| 36M| | 130 (2)| 00:00:01 | Q1,08 | PCWC | |
| 40 | INDEX FAST FULL SCAN | PK_DWH_DEBTOR | 6329K| 36M| | 130 (2)| 00:00:01 | Q1,08 | PCWP | |
| 41 | PX RECEIVE | | 6312K| 644M| | 5259 (1)| 00:00:01 | Q1,10 | PCWP | |
| 42 | PX SEND HASH | :TQ10009 | 6312K| 644M| | 5259 (1)| 00:00:01 | Q1,09 | P->P | HASH |
|* 43 | HASH JOIN RIGHT OUTER BUFFERED| | 6312K| 644M| | 5259 (1)| 00:00:01 | Q1,09 | PCWP | |
| 44 | PX RECEIVE | | 3689K| 31M| | 4271 (1)| 00:00:01 | Q1,09 | PCWP | |
| 45 | PX SEND HASH | :TQ10005 | 3689K| 31M| | 4271 (1)| 00:00:01 | Q1,05 | P->P | HASH |
| 46 | VIEW | | 3689K| 31M| | 4271 (1)| 00:00:01 | Q1,05 | PCWP | |
| 47 | HASH GROUP BY | | 3689K| 56M| 84M| 4271 (1)| 00:00:01 | Q1,05 | PCWP | |
| 48 | PX RECEIVE | | 3689K| 56M| | 3653 (1)| 00:00:01 | Q1,05 | PCWP | |
| 49 | PX SEND HASH | :TQ10003 | 3689K| 56M| | 3653 (1)| 00:00:01 | Q1,03 | P->P | HASH |
|* 50 | HASH JOIN | | 3689K| 56M| | 3653 (1)| 00:00:01 | Q1,03 | PCWP | |
| 51 | BUFFER SORT | | | | | | | Q1,03 | PCWC | |
| 52 | PX RECEIVE | | 3 | 21 | | 1 (0)| 00:00:01 | Q1,03 | PCWP | |
| 53 | PX SEND BROADCAST | :TQ10000 | 3 | 21 | | 1 (0)| 00:00:01 | | S->P | BROADCAST |
| 54 | INLIST ITERATOR | | | | | | | | | |
|* 55 | INDEX RANGE SCAN | I_DWH_PAYMENT_TYPE_EIDIID | 3 | 21 | | 1 (0)| 00:00:01 | | | |
| 56 | PX BLOCK ITERATOR | | 28M| 242M| | 3648 (1)| 00:00:01 | Q1,03 | PCWC | |
|* 57 | TABLE ACCESS FULL | DWH$PAYMENT | 28M| 242M| | 3648 (1)| 00:00:01 | Q1,03 | PCWP | |
| 58 | PX RECEIVE | | 6312K| 589M| | 986 (2)| 00:00:01 | Q1,09 | PCWP | |
| 59 | PX SEND HASH | :TQ10006 | 6312K| 589M| | 986 (2)| 00:00:01 | Q1,06 | P->P | HASH |
|* 60 | HASH JOIN | | 6312K| 589M| | 986 (2)| 00:00:01 | Q1,06 | PCWP | |
| 61 | PX RECEIVE | | 2937 | 172K| | 5 (20)| 00:00:01 | Q1,06 | PCWP | |
| 62 | PX SEND BROADCAST | :TQ10004 | 2937 | 172K| | 5 (20)| 00:00:01 | Q1,04 | P->P | BROADCAST |
|* 63 | HASH JOIN BUFFERED | | 2937 | 172K| | 5 (20)| 00:00:01 | Q1,04 | PCWP | |
| 64 | PX RECEIVE | | 220 | 1540 | | 2 (0)| 00:00:01 | Q1,04 | PCWP | |
| 65 | PX SEND HASH | :TQ10001 | 220 | 1540 | | 2 (0)| 00:00:01 | Q1,01 | P->P | HASH |
| 66 | PX BLOCK ITERATOR | | 220 | 1540 | | 2 (0)| 00:00:01 | Q1,01 | PCWC | |
| 67 | TABLE ACCESS FULL | DWH$MANDATOR | 220 | 1540 | | 2 (0)| 00:00:01 | Q1,01 | PCWP | |
| 68 | PX RECEIVE | | 2937 | 152K| | 2 (0)| 00:00:01 | Q1,04 | PCWP | |
| 69 | PX SEND HASH | :TQ10002 | 2937 | 152K| | 2 (0)| 00:00:01 | Q1,02 | P->P | HASH |
| 70 | PX BLOCK ITERATOR | | 2937 | 152K| | 2 (0)| 00:00:01 | Q1,02 | PCWC | |
| 71 | TABLE ACCESS FULL | DWH$PACKAGE | 2937 | 152K| | 2 (0)| 00:00:01 | Q1,02 | PCWP | |
| 72 | PX BLOCK ITERATOR | | 6312K| 228M| | 980 (1)| 00:00:01 | Q1,06 | PCWC | |
| 73 | TABLE ACCESS FULL | DWH$CASE | 6312K| 228M| | 980 (1)| 00:00:01 | Q1,06 | PCWP | |
| 74 | PX RECEIVE | | 78M| 4199M| | 10016 (1)| 00:00:01 | Q1,13 | PCWP | |
| 75 | PX SEND HASH | :TQ10011 | 78M| 4199M| | 10016 (1)| 00:00:01 | Q1,11 | P->P | HASH |
| 76 | PX BLOCK ITERATOR | | 78M| 4199M| | 10016 (1)| 00:00:01 | Q1,11 | PCWC | |
|* 77 | TABLE ACCESS FULL | DWH$ACTION | 78M| 4199M| | 10016 (1)| 00:00:01 | Q1,11 | PCWP | |
------------------------------------------------------------------------------------------------------------------------------------------------------------------

How to run multiple CodedUI Ordered Tests over multiple Test Agents for parallel execution using Test Controller

we are using VS 2013, I need to run multiple Coded UI Ordered Tests in parallel on different agents.
My requirement :
Example: I have 40 Coded UI Test scripts in single solution/project. i want to run in different OS environments(example 5 OS ). I have created 5 Ordered tests with the same 40 test cases.
I have one Controller machine and 5 test agent machines. Now I want my tests to be distributed in a way that every agent gets 1 Ordered test to execute.
Machine_C = Controller (Controls Machine_1,2,3,4,5)
Machine_1 = Test Agent 1 (Should execute Ordered Test 1 (ex: OS - WIN 7) )
Machine_2 = Test Agent 2 (Should execute Ordered Test 2 (ex:
OS - WIN 8) )
Machine_3 = Test Agent 3 (Should execute Ordered Test 3
(ex: OS - WIN 2008 server) )
Machine_4 = Test Agent 4 (Should execute Ordered Test 4 (ex:
OS - WIN 2012 server) )
Machine_5 = Test Agent 5 (Should execute Ordered Test 5 (ex:
OS - WIN 2003 server) )
I have changed the “MinimumTestsPerAgent” app setting value
as '1' in controller’s configuration file (QTController.exe.config).
When I run the Ordered tests from the test explorer all Test agent running with each Ordered test and showing the status as running. but with in the 5 Test Agents only 2 Agents executing the test cases remaining all 3 agents not executing the test cases but
status showing as 'running' still for long time (exp: More then 3 hr) after that all so its not responding.
I need to know how I can configure my controller or how I can tell it to execute these tests in parallel on different test agents. This will help me reducing the script execution time.
I am not sure what steps I am missing.
It will be of great help if someone can guide me how this can be achieved.
-- > One more thing Can I Run one Coded UI Ordered Test on One Specific Test Agent?
ex: Need to run ordered Test 1 in Win 7 OS (Test Agent 1) only.
Thanks in Advance.

Hi Divakar,
Thank you for posting in MSDN forum.
As far as I know, we cannot specify coded UI ordered test run on specific test agent. And it is mainly that test controller determine which coded UI ordered test assign to which test agent.
Generally, I know that if we want to run multiple CodedUI Ordered Tests over multiple Test Agents for parallel execution using Test Controller.
We will need to change the MinimumTestsPerAgent property to 1 in the test controller configuration file (QTControllerConfig.exe.config) as you said.
And then we will need to change the bucketSize number of tests/number of machines in the test settings.
For more information about how to set this bucketSize value, please refer the following blog.
http://blogs.msdn.com/b/aseemb/archive/2010/08/11/how-to-run-automated-tests-on-different-machines-in-parallel.aspx
You can refer this Jack's suggestion to run your coded UI ordered test in lab Environment or load test.
https://social.msdn.microsoft.com/Forums/vstudio/en-US/661e73da-5a08-4c9b-8e5a-fc08c5962783/run-different-codedui-tests-simultaneously-on-different-test-agents-from-a-single-test-controller?forum=vstest
Best Regards,
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

Oracle 8i hash parallel update question

Hello All,
Fairly new to Oracle (old Sybase guy). Using Oracle 8i. Trying to
get an update to run in parallel against a table. Partitioned the
table 4 ways using hash partitioining. Please see the following
query/plan. I am trying to determine if this query is actually running
in parallel or not from the plan? If it is not, what might be
preventing it from running in parallel? I would expect Oracle would
kick off one thread to update each of the four partitions of the
sales_tran_detail table.
Appreciated,
Jeff
UPDATE /*+ PARALLEL(sd, 4) */ SALES_TRAN_DETAIL_HASH SD
SET PRI_PUBLISHER_CALC_STATUS_CD = DECODE(9, 1,
PUBLISHER_CALC_STATUS_CD, 4, PUBLISHER_CALC_STATUS_CD, 7,
PUBLISHER_CALC_STATUS_CD, 10, PUBLISHER_CALC_STATUS_CD,
PRI_PUBLISHER_CALC_STATUS_CD),
PUBLISHER_CALC_STATUS_CD = '10', --If Monthly/Quarterly, then
'DROPOUT'
PUBLISHER_CALC_STAT_CHANGE_DT = '30-SEP-2006',
PUBLISHER_PROCESS_PERIOD_DT = '30-SEP-2006'
WHERE PUBLISHER_CALC_STATUS_CD IN ('09', '10', '30', '50')
AND EXISTS (SELECT 1
FROM SALES_TRAN_HEADER SH,
MCEF_INPUT_RPUPSALE MCEF
WHERE SH.SALES_PERIOD_DT <= '30-SEP-2006'
AND SD.SALES_PERIOD_DT = SH.SALES_PERIOD_DT
AND SD.BATCH_ID = SH.BATCH_ID
AND SD.AFFILIATE_ID = SH.AFFILIATE_ID
AND MCEF.DATAFILE_CD = 'D' -- Records in Error
AND MCEF.BATCH_ID IN (' ', SH.BATCH_ID)
AND MCEF.ALT_COPYRIGHT_IN = 'N'
AND SH.DISTRIBUTION_METHOD_CD =
MCEF.DISTRIBUTION_METHOD_CD
AND SH.AFFILIATE_ID = MCEF.AFFILIATE_ID
AND SD.AFFILIATE_ID = MCEF.AFFILIATE_ID
AND SD.CATLG_CORE_ID = MCEF.CATLG_CORE_ID
AND SD.CATLG_PFIX_ID = MCEF.CATLG_PFIX_ID)
EXPLAIN PLAN from QUEST:
1 Every row in the table Partitions accessed #1 - #4 is read.
2 All partitions of steps 1 were accessed..
3 Rows were retrieved using the unique index PRD1.SALES_TRAN_HEADER_P.
4 Rows from table PRD1.SALES_TRAN_HEADER were accessed using rowid got
from an index.
5 One or more rows were retrieved using index
PRD1.MCEF_INPUT_RPUPSALE_N1. The index was scanned in ascending order.
6 For each row retrieved by step 4, the operation in step 5 was
performed to find a matching row.
7 For the rows returned by step 6, filter out rows depending on filter
criteria.
8 For the rows returned by step 2, filter out rows depending on filter
criteria.
9 Rows were updated.
10 Rows were updated.

Thanks everyone for the responses on this. I have tried everything suggested thus far and still do not get what I can recognize as an explain plan that shows parallel execution or multiple threads when running. I would have thought the hint would have forced it to happen!
My parallel_max_servers is set to 6.
I copied in the header as stated in goodwithalls email.
I enabled parallel dml for the session and played with "..adaptive_multiuser_tuning".
I still get exactly the same plan as posted and still 1 thread. I wonder if Oracle just doesn't like doing updates vs. hash paritioned tables in 8i? It claims it does it vs. paritioned tables but maybe hash doesn't count.. or it wants me to create paritioned indexes too or something wierd..
Still lost on this one,
Any other thoughts anyone?
Thanks,
Jeff

Parallel execution in a web template

We have a web template with several queries. It is possible execute our queries with parallel execution?
Thanks !!

Hi Dani,
currently all DP's are executed sequentially, because in general it would be possible that the dataproviders have influence on each other.
i heard that sap is building a prototyp to allow parallelization in BEX web 7.0. But currently there is no release date known, May this will come with some enhancement package.
You should consider the Functional enhancement schedule on http://service.sap.com/bi. If this feature will be released sometime, it will be listed there.
best regards,
Kai

Parallel execution of multiple test cases.

Does flex unit 4.x version supports for parallel execution of multiple test cases?
Any help will be greatly appreciated.

No.
Flash Player and AIR are single-threaded virtual machines. The only parallel execution one could do is during asynchronous tests, but we do not support that presently for a variety of reasons.
Mike

Parallel Execution against Normal Execution

Hi,
Can someone explain why in this case Serial Execution is faster although enabling parallel dml at session level , What are the possibilities to improve speed of executoin for parallel execution as i need to populate for 139 stores data having around half a million rows for each store , i will have to run a cursor on the distinct value of store to achieve this , just to test i got the following Results , can anyoe guide where i am going Wrong?
This is the Script
set timing on
insert into r8_win_store
select * from win_store@rms8 where store=99;
commit;
alter session enable parallel dml;
insert /*+ Parallel(t,8) */ into r8_win_store t
select /*+ parallel(e,8)*/ * from win_store@rms8 e where e.store=99;
commit;
alter session disable parallel dml;
OUtput
SQL> @test1.sql
299666 rows created.
Elapsed: 00:03:48.12
Commit complete.
Elapsed: 00:00:00.01
Session altered.
Elapsed: 00:00:00.00
299666 rows created.
Elapsed: 00:08:02.81
Commit complete.
Elapsed: 00:00:01.31
Session altered.
Elapsed: 00:00:00.00

Parallel processing in Oracle is intended to reduce I/O latency. When you tell the kernel to do an I/O, you need to wait for that call to complete - disk platters to spin, disk controller heads to move, etc.
If you need to make a bunch of I/O calls after one another, you will spend up to 90% of the elapsed time waiting for that I/O to actually put the data into your buffer to process.
Now imagine doing a million I/Os.. and 90% of the time spend waiting for I/O to complete. A big performance knock and a frustrating one - as you cannot make that actual I/O any faster. It is simply slow. The slowest operation you can do on the computer and you're doing a million of them.
Parallel processing in Oracle addresses this problem. There can be sufficient I/O bandwidth to make a 100 I/O calls per second. But your process that does an I/O call, wait, process, and does an I/O call, wait, process, can only reach a speed of 10 I/O calls per second.
Which means 90% of the I/O pipe is free to do I/O in parallel. So instead of a single process doing that million I/Os (using 10% I/O bandwidth/thruput), Oracle spawns 10 process each doing a 100K I/Os each - thus making better use of the I/O thruput.
So Oracle PQ is useless if you scan small volumes of data - it is intended for large volumes of data. There are overheads in PQ as the parallel processes have to be coordinated in order to work together. When each only needs to do 10 I/Os, the time spend on coordination alone can be more than what the time would have been to simply do a 100 I/Os using a single process.
It is also a fallacy that number of CPUs determine the just how many parallel processes you can start. The real determining factor is the load you can put on your I/O subsystem.
Bottom line is that PQ is nothing "special" or "magic". It is simply a method to reduce I/O latency by performing parallel I/O. And it is only sensible to use when the amount of I/O to be done warrants parallel processing.
Oh yeah - and the CBO is very capable in deciding when to use PQ and not. So rather than force PQ down the CBO's throat using the PARALLEL clause and hardcoding the degrees and instances, it should rather make those decisions (as informed decisions) itself. Which means using the PARALLEL clause on tables and not as SQL hints. The DBA can easily tune that by altering the PARALLEL clause if a table.. The DBA cannot by any means do the same thing when dealing with SQLs that insists on a specific number of PQ processes on a specific number of instances.

Parallel execution for Expression Filters

I'm looking for any information on whether expression filters are amenable to parallel execution or not.
I had a look through the documentation without success, and saw that in the type declarations for supporting objects there is no mention of DETERMINISTIC or PARALLEL_ENABLE. However, the static functions and procedures do not look like they have any constructs that the database would see as an impediment for parallel execution.
Can anyone advise on this?

Thanks for that.
Maybe in a parallel query environment the index will not be used. But I was really wondering whether the execution of the Evaluate function itself would be parallelised, as it invokes functions that are not explicitly marked as deterministic or as safe for parallel execution.
http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14251/adfns_packages.htm#BABIGDGE
Well, I'll see if I can investigate (maybe I'll have to suspend those system triggers that prevent modification of the Type ... just for experimental reasons, you understand).

Parallel Execution questions

Similar Messages

Maybe you are looking for