Query Explain without running it.

I use trans code rsrt to display the execution plan of my slow running query. Can anyone tell me how to do this WITHOUT running the query. Each time I try and exit it runs.
Thanks

Hi Richard,
I am not so sure but I think any optimizer program needs to run the QUERY somehow to detect the execution plan. Even if you go straight to the RDBMS (<i>e.g. MSSQL, ORACLE etc.</i>) bundled <i>optimizer/analyzer utilities</i> with them somehow executes the <b>SQL Query</b> to detect <i>bottlenecks</i> and problematic sections of the SQL Query (w/c is of no difference to BEX Query as in your Case). 
Regards,
--Jkyle

Similar Messages

  • How i can obtain explain plan without run the query

    Hello,
    i need to know the result of explain of a query without run the query, it's this possible?
    Thanks and best regards.

    explain plan for <your query>;
    select * from table(dbms_xplan.display);Regards,
    Rob.

  • How to get query execution time without running...?

    Hi ,
    I had one requirement .... as follows ......
    i had 3 sql statements . I need to execute only one sql which execution time is very less.
    Can any one help me , how to get query execution time without running that query and without using explain plan..?
    Thanks,
    Rajesh

    Kim Berg Hansen wrote:
    But you have ruled out explain plan for some reason, so I cannot help you.OP might get some answers if query was executed before - but since restart. Check V$SQL dynamic performance view for SQL_TEXT = your query. Then ROUND(ELAPSED_TIME / EXECUTIONS / 1000000) will give you average elapsed time.
    SY.
    Edited by: Solomon Yakobson on Apr 3, 2012 8:44 AM

  • Get est time for sql without running against the db ??? possible ?

    hello,
    is there anyways to get a query execution time without running that against the database ?? even an estimate would be good...i am on 10.2.0.3 ......i am it might be impossible for it...but anyways to get estimate besides setting autotrace only ?? as i said want nothing but just like a estimate time for how long will it take to finish the query...i know we can very long_ops view but as i said..without running against the DB ...is that possible ?? as the explain plan i get shows time ?? but is that how long it will take to finish up or ?/
    select * from table(dbms_xplan.display);
    | Id  | Operation            | Name | Rows  | Bytes | Cost (%CPU)| Time     |  
    |   0 | SELECT STATEMENT     |      |    56 |  3192 |     9   (0)| 00:00:01 |  
    |   1 |  MERGE JOIN CARTESIAN|      |    56 |  3192 |     9   (0)| 00:00:01 |  
    |   2 |   TABLE ACCESS FULL  | DEPT |     4 |    80 |     3   (0)| 00:00:01 |  
    |   3 |   BUFFER SORT        |      |    14 |   518 |     6   (0)| 00:00:01 |  
    |   4 |    TABLE ACCESS FULL | EMP  |    14 |   518 |     2   (0)| 00:00:01 |  
    -----------------------------------------------------------------------------  

    user11168115 wrote:
    hello,
    is there anyways to get a query execution time without running that against the database ?? <snip>
    How long is a string?
    How deep is a hole?
    How long does it take to read an unknown number of blocks containing an unknown number of rows from an unknown disk system with an unknown contention load, pulling it across an network of unknown bandwidth and unknown load ....
    And I'm not saying that if you somehow provide all this that the answer can be known. I'm saying the too much of the information tht would be needed to make such an estimate is unknowable. Much of it will be widely variable from one execution of the query to the next execution of the exact same query. Which leaves you with exactly what to base your estimate on?
    Without even seeing your query, I can tell you with absolute certainty that it will take between 1 nano-second and 100 years to execute.

  • Query is not running after one table exceeds certain records, any limit?

    I have a production database and a reporting database. Production is on version 9.2.0.5 where as Reporting is on 9.2.0.3. I had a query which uses Gl table which grows every month. The query use to take almost same time both in Production as well as Reporting. After the table grew in last couple of months. I observed the query just hangs up in production where as it continues to run in Reporting. Is it the orcale version which makes any difference for the query not to run in prodcution when Gl table had 27 million rows as against 25 million rows a couple of months of ago? Any help is appreciated.

    user5846372 wrote:
    If I use the gl table which was populated until two months ago having less than say 25.2 million rows, the query runs in 2 minutes. If I use the current gl table having current data (total of say 27 million rows) then the query does not even run. From the current gl table if I delete about 2 million records making it 25 illkion rows then the query runs. So I was wondering whether it is the oracle version which makes this difference.Explain your logic that says that after the queries run successfully on different versions, the version difference is now somehow causing the problem - and not the fact that what did change since the last successful run was the the data volumes....
    As for "+the query does not even run+" - it does. Simply that you do not have a clue what it is doing and thus no means to diagnose the problem. And then jump to the conclusion that the version difference is somehow to blame.
    WHAT is that query doing when it seems to "hang"? (it very likely does not hang at all, and is simply very busy with I/O or being blocked from accessing a resource it needs)
    Have you looked at the events and wait states of the query? Have you looked, as Daniel already suggested, at the execution plan?
    You cannot and should not attempt to diagnose a problem, without understanding the problem. If you have no idea what the problem is, how can you simply jump to the conclusion that "+oh, it must be the version difference+"!?
    You can only solve a problem by understanding it. It is that simple. So first understand what the query is doing and why it seems to hang BEFORE you attempt to diagnose and solve it.

  • Have a month number but QaaWs query will not run

    If I type in the parameters for the query it runs with no problem, but when I use the MONTH function that returns the month number the query will not run.
    I think the date serial number is what is being passed even though I see a number being display.  Does anyone know how to conver this MONTH() function to a true number?  This is a converted 4.5 model that works just fine, but looks like they tightened things up in 2008.
    Thanks...
    Tom Nather

    Here is how its setup.  I have a QaaWs query that returns the max date, then Excel functions that calculate the date back 13 months.  One of the functions is MONTH() that returns the month number based on a date stored in a cell, not really a true number.
    I added a vlookup that takes the month from a date field (05/31/2008) and looks up the appropriate NUMBER of the month =VLOOKUP(MONTH(K47),M54:M65,1). Find 5 (which is defined as a number.
    This then is applied to the input side for the Qaaws (Month Number = double) using a Paste Link function from the CONTROLS sheet in excel, and the query then works.
    Anything that is a date always seems to be built off of the date serial number so this was the easiest way to get this to work. 
    I think standards have been tightened up using the Flex language vs. what 4.5 was written in, this works in 4.5 without the necessary conversion.

  • Could someone please explain what "Run Spindump" & "Run System Diagnostics" does in 'Activity Monitor' ?

    Could someone please explain what "Run Spindump" & "Run System Diagnostics" does in 'Activity Monitor' ?

    Good question.  You don't need to worry about running those.  They will run automatically when needed by the system.  The information they generate (some) is sent to apple automatically (unless that setting is disabled).
    If you want to look at the output of these files, most can be found in the console (Utility/Console).  Open the app and look at all the system messages your Mac generates....on the right hand side I would bet you already have a crash log or spindump that was collected without you even knowing it.

  • I have a Macbook pro 2009 with Snow Leopard.  Which system can I install without running into complicated problems which I will not be able to solveby myself? (Mountain Lion, Mavericks, Yosemite?)

    I have a Macbook pro 2009 with Snow Leopard.  Which system can I install without running into complicated problems which I will not be able to solve by myself- a person with limited problem solving abilities. (Mountain Lion, Mavericks, Yosemite?)

    It will make it easier to help you w/ your problem to know the  size and RAM installed.
    "Yesterday I couldn't update the maps on my Garmin because it said the OS I had wouldn't work"
    What OS will your Garmin work with? Funny there was someone else that did that and upgraded to Yosemite and regretted it.
    I'd say Lion if your Garmin is compatible.
    Both Lion and Mountain Lion are downloadable from the Apple Store:
    http://store.apple.com/us/product/D6106Z/A/os-x-lion
    http://store.apple.com/us/product/D6377Z/A/os-x-mountain-lion
    The OSs will get more resource hungry the higher you go.

  • HT201263 my ipad give me the apple sign without running up it appears all the time in the screen , i tries to reset my ipad , but still in this problem , can you please help me?

    my ipad give me the apple sign without running up it appears all the time in the screen , i tries to reset my ipad , but still in this problem , can you please help me?

    You may want to look at this about using recovery mode to restore your iPad.
    http://support.apple.com/kb/ht4097

  • Is it possible to install both FCP7 and FCP X on your hard drive without running into problems?

    Is it possible to install both FCP7 and FCP X on your hard drive without running into problems?

    Apple recommends separate partitions: http://support.apple.com/kb/HT4722

  • F110 run the payment run directly without running the proposal run first

    Hello to you all,
    Does any one know I do I block the option to run the payment run directly without running the proposal run first?
    Thanks in advance,
    Yoav

    Hi
    First delete the payment run output by the following :
    Goto Edit on menu bar --> Payment --> Delete Output
    then run the proposal run again and after that process payment run.

  • Stream-in images from client to server,without running the transmit code

    Hello all,
    Im using JMF for my current project.Right now,I am trying to only recieve images from the client without the client transmitting it.That is I dont want the transmit code to run on the client.
    I should just be able to stream in the images from client to server, without running the transmit code on client.
    Can this be done?
    Thanks in advance

    suppigs wrote:
    Can I know more about this?Sure.
    <Side A>
    You'd just need to write an application that doesn't have a GUI (so, a console-based application) that listens on some pre-determined port for a message to start broadcasting. Maybe you'd send it the IP/PORT number to start broadcasting on. Once it receives that message, it'd start broadcasting the web cam to the IP/PORT number until it received a message to stop. Once it stops, it'll just go back to waiting for the next "start" signal.
    <Side B>
    On the other side, you'd write an application that sends the start messages, receives/displays the videos, and then sends the stop signal. This will have a GUI, and be your "control program" so to speak.
    Then, once of have both of those programs working...if you're using Linux, you're done. If you're using Windows, you'd need to modify the <Side A> program so that it can run as a Windows service.
    There are a lot of ways to do this, you can google it or look at the following link:
    [http://twit88.com/blog/2007/09/19/open-source-software-to-start-up-java-as-windows-serviceunix-daemon/]

  • FRM - 41105 You Cannot query records without a saved parent record

    I have 3 data blocks. The contro data block have document type and document number if we come to new record(I have written execute query for both header and detail part) the header block will show the detail of the documnet header(Header Details) and detail block will show the detail part.
    But my problem is when new record instance only the header block is querying and detail block is not querying and giving an error message(*FRM - 41105 You Cannot query records without a saved parent record*)!!
    Plz help me!!
    Thanks in advance,

    yea header tables PK is refered in the details table.
    and i have master record created already and have the detail record for those master records in the table.
    i want to query the detail record related to that master when we select a document number and document code in the control block.
    Its querying the master detail relevent to that header and giving this error messge while quering the detail record..

  • Run SSIS package at scheduled time without running SSIS service

    Hello
    I want to run SSIS Package using SQL Agent job, But for some reason, I am allowed to install SSIS services on the server. Is there any way to run SSIS package without running SSIS service? I tried using DTEXEC.exe command, but it still give an error. "To
    run SSIS package outside SQL server data tools you must install service".
    Thanks in advance

    Hi Swetha,
    The service itself is not doing much (only allows to connect), but to actually execute SSIS packages naturally you have to have the Integration Services installed where you want to run the package. Likewise, if you have SSIS installed elsewhere you can run
    it remotely off the local SQL Agent that connects to a remote package.
    Arthur My Blog

  • How to populate one array without running the entire VI

    I'm trying to populate a complex array that I built that reads from a spreadsheet file.  The company I am building it for only has labview run time on the PXI test equipment that I am working with.  I am having to fix the VI On my computer then build an executable and then test it on the machine to see if it runs properly.  It's a very tedious process.  I cannot run the VI on my computer because it requires to have a FLUKE 726 connected in order to run.  So, I was wondering if there is a way to repopulate the array without running the entire VI?
    Your help would be greatly appreciated.

    Create a subVI for reading/processing of the array.  Then you just have to test that subVI as your machine.
    There are only two ways to tell somebody thanks: Kudos and Marked Solutions
    Unofficial Forum Rules and Guidelines

Maybe you are looking for