Cube operator and aggregation of measures on existing records

Hi there,
I'm using a cube operator with loading type LOAD in order to perform merge on fact table.
We have following situation
Record already exists on fact table for customer and product with qty 8.
Incoming record has qty 2, so tried using cube operator hoping that because aggragtion specfied on qty measure was sum and set to solve cube, new qty would be 10.
However, looked at sql generated and record simply gets updated with new qty rather than adding new to existing qty.
I can achieve our aim by simply reading for any existing record and adding new record qty to existing record qty, but was hoping the cube operator would do thi s for me.
Anybody achivede anything similar using sinmply cube operator.
Many Thanks

Do you mean that you want to load the data into the AW using the AVG function instead of SUM? If this is true, are you planning to use AVG as the aggregation operator in the AW as well? Will this give the answer you want? The code currently defaults to SUM for load even if you aggregate the cube using AVG since AVG of AVG is not usually what people want. If you want to do it anyway then there it is possible if you hand edit the XML to add an attribute named AggregationMethod to the CubeMap. E.g.
    <CubeMap
      Name="MAP1"
      Query="SALES_FACT"
      AggregationMethod="AVG">But the simpler way to do it is to define a SQL View that aggregates to the load level using AVG and then map the cube to the view.

Similar Messages

  • Answers - Columns and aggregation of measures

    Can I bring in data at a line item (Order_ID) level in Answers do some calculations around a metric returned, then be able to hide/remove this column in a table and have the metrics be aggregated up to the other level - say if I bought in Region? I have some bucketing to do around an Order metric but don't want to present this granularity in the table.
    Thanks

    Yvan.
    In the end I settled for aggregating tax year data (for fifteen years) as this seems to be the main reason for being retrospective in my accounting. I had to reference this manually in the ledger.
    I then used formula similar to this in the crosstab:
    =SUMIFS(Register :: $gbpe,Register :: $Field,"=Income",Register :: $Type,"=Salary",Register :: $Tax,$A11) where $A11 refers to the crosstab year row in question.
    I see a problem with the computing load these formulae have created, but am pleased to be able to do the analysis, as I was reliant on pivot tables in Excel and wondered how I could cope in Numbers.
    Thanks for your help

  • Cube Compression and Aggregation

    Hello BW Gurus,
    Can I first compress my infocube data and load data into the aggregates.
    The reason being, that when the infocube is compressed the Request Id's are removed.
    Are the Request Id's necessary for data to be transfered to the Aggregates and then later on for aggregate compression.
    Kindly suggest.
    regards,
    TR PRADEEP

    Hi,
    just to clarify this:
    1) you can compress your infocube and then INITIALLY fill the aggregates. The Request information is then no longer needed.
    2) But you can NOT compress requests in your infocube, when your aggregates are already filled and these requests are not yet "rolled up" into the aggregates (this action is prohibited anyway by the system).
    Hope this helps,
    Klaus

  • Using dimension and cube operator

    hi
    I am facing problem in using this dimension and cube operator..
    Actually I want to create a sales data with dimensions as time,customer and product
    After creating a dimension in owb I get a table created attached with it same is the case with cube operator
    so where should I store my measures in the cube or the table created and where to store the data about dimensions in the table created.
    and finally how to use them in mapping to retrieve the data stored
    Thanks

    It sounds like you are not getting any matching keys for loading into the cube.
    Do you have a time dimension created by OWB in the cube? The key used by the cube operator for an OWB time dimension is a formatted number. The time dimension keys are stored as follows;
    Day Level - YYYYMMDD
    Month Level - YYYYMM
    Week Level - YYYYWW
    Quarter - YYYYQ
    Year - YYYY
    If you have a source that has a SQL date datatype for example and want to construct the key for a cube's time dimension at the day level something like the following expression can be used to construct the time reference from a SQL date...
    to_number(to_char( time_key, 'YYYYMMDD'))
    It may not be this but just a thought.
    Cheers
    David

  • I'm trying to clean up my mailbox and when I delete an item(s) I get a message that say's "operation could not be completed- file exists. Further, when I do this one message is copied (not deleted) to the trash folder. Can someone help me through this?

    I'm trying to clean my mailbox and when I click on delete or highlight and move to trash I get a message that says "operation could not be completed - file exists. Further, when I select a message or group of messages, one of them is copied (not deleted) to the trash folder. Can someone help me through this?
    Thanks.

    It's a POP account. I decided to clean up my mailbox today after noticing the computer was real slow after USA Today started sending their daily top stories with new graphics. Also before I noticed the problem with mail deletion - two things happened. 1) I updated security sofware as prompted by Apple and 2) downloaded Mackeeper and went throught some preliminary scans but did not buy their package. Thanks forn nstaying with me.

  • Doing Data acq. and buffered period measurement using counters simultaneously, gives an buffer overflow error

    Iam doing Data acquisition using NI-PXI 4472 and buffered period Measurement using NI-PXI 6602 simultaneously,my program gives an buffer overflow error

    murali_vml,
    There are two common buffer overflow and overwrite errors.
    Overflow error -10845 occurs when the NI-DAQ driver cannot read data from the DAQ device's FIFO buffer fast enough to keep up with the acquired data as it flows to the buffer (i.e., the FIFO buffer overflows before all the original data can be read from it). This is usually due to limitations of your computer system, most commonly the result of slow processor speeds (< 200 MHz) in conjunction with PCMCIA DAQ boards, which have small FIFO buffers (e.g., the DAQCard-500). Sometimes using a DAQCard with a larger FIFO can solve the problem, but a better solution is to lower the acquisition rate or move to a faster system. Another cause of the -10845 error could be due to an interrupt-driven
    acquisition. For example, PCMCIA bus does not support Direct Memory Access (DMA). If the system is tied up processing another interrupt (like performing a screen refresh or responding to a mouse movement) when it is time to move data from the board, then that data may get overwritten.
    Overwrite error -10846 occurs when the data in the software buffer that you created for an analog input operation gets overwritten by new data before you can retrieve the existing data from the buffer. This problem can be solved by adjusting the parameters of your data acquisition, such as the lowering the scan rate, increasing the buffer size, and/or increasing the number of scans to read from the buffer on each buffer read. Additionally, performing less processing in the loop can help avoid the -10846 error.
    See the NI-DAQ Function Reference Manual for a listing of all NI-DAQ error codes.
    Have a great day.

  • Data in the Cube not getting aggregated

    Hi Friends
    We have Cube 1 and Cube 2.
    The data flow is represented below:
    R/3 DataSource>Cube1>Cube2
    In Cube1 data is Stored according to the Calender Day.
    Cube2 has Calweek.
    In Transformations of Cube 1 and Cube 2 Calday of Cube 1 is mapped to Calweek of Cube 2.
    In the CUBE2 when i upload data from Cube1.Keyfigure Values are not getting summed.
    EXAMPLE: Data in Cube 1
    MatNo CustNo qty calday
    10001  xyz     100  01.01.2010
    10001  xyz      100  02.01.2010
    10001  xyz      100   03.01.2010
    10001  xyz     100  04.01.2010
    10001  xyz      100  05.01.2010
    10001  xyz      100   06.01.2010
    10001  xyz      100   07.01.2010
    Data in Cube 2:
    MatNo CustNo qty calweek
    10001  xyz     100  01.2010
    10001  xyz      100  01.2010
    10001  xyz      100   01.2010
    10001  xyz     100   01.2010
    10001  xyz      100   01.2010
    10001  xyz      100   01.2010
    10001  xyz      100   01.2010
    But Expected Output Should be:
    MatNo CustNo qty calweek
    10001  xyz     700  01.2010
    How to acheive this?
    I checked in the transformations all keyfigures are maintained in aggregation summation
    regards
    Preetam

    Just now i performed consisyency check for the cube:
    I a getting following warnings:
    Time characteristic 0CALWEEK value 200915 does not fit with time char 0CALMONTH val 0
    Consistency of time dimension of InfoCube &1
    Description
    This test checks whether or not the time characteristics of the InfoCube used in the time dimension are consistent. The consistency of time characteristics is extremely important for non-cumulative Cubes and partitioned InfoCubes.
    Values that do not fit together in the time dimension of an InfoCube result in incorrect results for non-cumulative cubes and InfoCubes that are partitioned according to time characteristics.
    For InfoCubes that have been partitioned according to time characteristics, conditions for the partitioning characteristic are derived from restrictions for the time characteristic.
    Errors
    When an error arises the InfoCube is marked as a Cube with an non-consistent time dimension. This has the following consequences:
    The derivation of conditions for partitioning criteria is deactivated on account of the non-fitting time characteristics. This usually has a negative effect on performance.
    When the InfoCube contains non-cumulatives, the system generates a warning for each query indicating that the displayed data may be incorrect.
    Repair Options
    Caution
    No action is required if the InfoCube does not contain non-cumulatives or is not partitioned.
    If the Infocube is partitioned, an action is only required if the read performance has gotten worse.
    You cannot automatically repair the entries of the time dimension table. However, you are able to delete entries that are no longer in use from the time dimension table.
    The system displays whether the incorrect dimension entries are still being used in the fact table.
    If these entries are no longer being used, you can carry out an automatic repair. In this case, all time dimension entries not being used in the fact table are removed.
    After the repair, the system checks whether or not the dimension is correct. If the time dimension is correct again, the InfoCube is marked as an InfoCube with a correct time dimension once again.
    If the entries are still being used, use transaction Listcube to check which data packages are affected.  You may be able to delete the data packages and then use the repair to remove the time dimension entries no longer being used. You can then reload the deleted data packages. Otherwise the InfoCube has to be built again.

  • Report with non aggregated and aggregated columns from different facts.

    Hi,
    We have got requirement as follows,
    1) We have two dimension tables, and two fact(Fact1 and Fact2) table in physical.
    2) In BMM we have made hierarchies for both dimensions, and are joins both logical fact table.
    3)In fact1, we are having three measures of which we have made two as aggregation sum, and one is non aggregated(It contains character).
    4)Fact2 have two measures, both are aggregation as sum.
    5)Now here the problem arises, we want to make a report with some columns from dim and non aggrgated column from fact1 and and aggregated column fact2
    How to resolve the above issue.
    Regards,
    Ankit

    As suggested you really want to move your none-aggregated fact attributes to a logical dimension (using the same physical table as the logical fact). Map this in the BMM layer as a snowflake, Place a hierarchy on this dimension with (at minimum) Total -> Detail levels, then on the other fact table you want to include in the report, set the content level on your other fact measures to the 'Total' level for your new logical Dim and it will allow them to be present in the same report.

  • Limiting Tables and charts with Measures having RelativeValue

    Hallo,
    I just try to do some difficult stuff with the WebI Rich Client 4.1.
    There are two measures for each month on a time line showing year and month. Thus, I have a dimension for year and month and two different measures for each year/month combination.
    Now, e.g. I want to compare these two measures having same month and different years, say the current year and the year before. Please see the table below for a visible figure.
    I used RelativeValue([measure prior year_1];([year/month]);-12) for [measure prior year] in order to compare one 1 year older measure with the other measure.
    year/month
    measure prior year
    measure current year
    2010/10
    2010/11
    2010/12
    2011/01
    2011/02
    2011/03
    2011/04
    2011/05
    2011/06
    2011/07
    2011/08
    2011/09
    2011/10
    439
    97
    2011/11
    7.294
    2.712
    2012/02
    20.496
    13.397
    2012/05
    21.859
    12.636
    2013/01
    36.325
    27.286
    2014/08
    26.894
    17.425
    There are no values for [measure current year] before 2011/10. But, there are values for [measure prior year_1] before 2011/10. [measure prior year] is actually from 2010/10, 2010/11 and so on, as defined with RelativeValue.
    No problem here. Everything works fine. But, now, I want to start the table with the first row where [measure prior year] is not null while leaving measure current year unchanged. Just cut the table and start with 2011/10...
    And here we get the problem. Hence, RelativeValue references to values before 2011/10 all these reference will disappear for [measure prior year].
    The same is with charts. I couldn't find a way to reference to values that exist by the query but which are not displayed in a table or a chart!
    What I need is a way to reference to values which are not present in a single table or chart but in the report. I already fiddled with the display options (null values, without dimension etc.) of tables. But, no way.
    How could a accomplish this task?
    Any idea of something like "hiding" unwanted rows/bars results in "loss of data".
    Thanks for any help
    Matthias

    Well... if you think and try long enough, you may find the solution .
    The core problem is to reference to queried values not shown in the table/chart. And there IS a solution.
    Just use...
    NoFilter(RelativeValue([measure prior year_1];([year/month]);-12))
    ... instead of
    RelativeValue([measure prior year_1];([year/month]);-12)
    NoFilter enables RelativeValue to reference to values not present in a table or chart, but available in the report .
    From there on, you can define e.g. the starting point of the category axis of a chart by defining a local filter without losing values to be referenced .

  • JDBC Scenario: Delete all the existing records and Insert

    Dear All,
    Scenario: IDoc to JDBC
    PI need to delete all the existing records in the SQL table then need to insert the records.I have created a two statements in the target structure.
    Statement1
       -> TableName
           -> action : DELETE
           -> Table : TableName
    Statement2
       -> TableName
           -> action : INSERT
           -> Table : TableName
           ->Access
               -> Elements
    I am getting below error while try to delete the records in the table.
    JDBC message processing failed; reason Error processing request in sax parser: Error when executing statement for  table/stored proc. 'Tablename' (structure 'Statement1'): java.sql.SQLException: FATAL ERROR document format: structure 'Statement1', no key element found
    Note: we use above structure in the another interface without specifying any Key for Delete operation(Server is same but different database for this two interface) and it works fine. Although the Primary keys has been set in the SQL table i am getting the above error.
    let me know what are the setting need to make in PI and SQL DB to achieve this.
    Regards,
    Manikandan Shanmugam
    Edited by: Manikandan Shanmugam on Nov 8, 2011 6:19 AM

    The straucture what you have deined for Delete is not correct, you should have key field in statement structure then only you can perform Delete operation.
    refer below link
    http://help.sap.com/saphelp_nwpi71/helpdata/EN/44/7b6e85fde93673e10000000a114a6b/content.htm
    Regards,
    Raj

  • Phases/operation in process order does not exist

    phases/operation in process order does not exist  this error coming when cor6n is using.why?

    HI
    After creating Master recipe in C201, under the operations tab you need to define the operations involved in recipe . along with the operation column  there is phase indicator column is there which is to be selected whether this operation is phase or not...
    Check and revert
    Regards
    Anupam Sharma

  • Updating Physical Cube Tables and Hierarchies OBIEE 11.1.1.7

    OBIEE 11.1.1.7
    I have imported a MSAS Cube, modeled it and created a presentation layer. I now have some new hierarchies I need to add to the phyicial layer.
    How do you import new Cube Tables and/or Hierarchies?
    When I select the Import option it creates a new OLAP database and connection pool, I am unable to successfully move those changes into the existing OLAP Connection.
    I have been through the documentation and it reccomends importing over a manual process. However I am unable to figure out how to do this with the import process.
    Oracle® Fusion Middleware Metadata Repository Builder's Guide for Oracle Business Intelligence Enterprise Edition
    11g Release 1 (11.1.1) Part Number E10540-05
    Chapter 8 Working with Physical Tables, Cubes, and Joins
    There is a section called Working with Multidimensional Sources in the Physical Layer, in this section is says
    "Each multidimensional catalog in the data source can contain multiple physical cubes. You can import the metadata for one or more of these cubes into your Oracle BI repository. Although it is possible to create a cube table manually, it is recommended that you import metadata for cube tables and their components"
    The New Utilities feature(s) is really cool, but it doesn't work for OLAP connections. 
    Any help is greatly appreciated.

    Hi Michael,
    it works, but its a bit tricky. I did it with Oracle OLAP and I think it must be similar with other OLAP Databases.
    Rename your existing Database in Physical Layer to the name Metadata Import dialog would create. With Oracle OLAP the "Data Source Name" ist used. Just try it once.
    Unfortunately you can not merge dimensions and cubes as it is with relational Objects . So you have to delete the cube and dimensions you will import from you renamed Database in Physical Layer. Of course the mapping between physical and business model will be lost
    Now you can import your modified OLAP-Cubes and Dimensions and it should be placed in your renamed Database in Physical Layer
    Open the sources of your logical tables in business model and add the Dimensions or Cubes in the "General" Tab. Move to the Tab "Column Mapping" and check if mapping is OK. If not (this will be the case, if you customized the column names), you have to do the mapping from the scratch .
    I recommend to use the original names from the data source. If you need another name for a column, just create a new logical column and use "Derive from existing columns". Now you can reimport OLAP Metadate quicker because the Mapping is done automatically.
    Hope this helps
    Regards Claus

  • How data update when there is two planning cubes in one aggregation level

    If we have two planning cubes and a multi-cube includes these two planning cubes.
    then we have the aggregation level defined based on this multi-cube.
    if the characteristics and key figures defined in aggregation level all are included in these two planning cubes. When we plan the data with plan query based on this aggregation level, which plan cube will be updated with plan data, or both of them?

    HI Wang,
    yes as Bindu said you need to restrict the layout based on the infoprovider.
    Lets assume that we have a multi provider M1 and it has two planning cubes cube1 and cube1.
    in case u develop a layout for updating cube 1 give the restriction in the filter section as infoprovider = cube1.
    or the vice versa.
    Else ur layout must input values for both the cubes simultaneously, then in that case create two restricted KF with each having a restriction for cube1 and cube 2.
    so wen the user enters value for the first restricted KF it will get saved in cube1 and wen he enters value for the second  restricted KF it will gets saved in cube2.
    This is how it works. Guess this was useful to u.
    Regards.
    Shafi.

  • Using cube operator in mapping

    Hi everyone,
    I am just building my first owb map with a cube (relational fact table) but I can't find any description anywhere on how to actually use/connect the cube operator in a mapping. I have managed to build a cube, linked to two dimensions, but when I place it on the canvas, I am not sure what it expects.
    Does anybody know of any good examples or a bit of documentation anywhere?
    Thanks!
    Ed
    PS In a SQL statement I would populate that fact table by simply joining the external table with the dimension, but the input signature of the cube is just different from what I expected.

    hi Patrick
    i have already done this but it is not working as well and response is same "you are using two different sources in expression".
    in fact i am using few tables from source schema and another table(time dimension table) from target schema. i think either this error is due to using time dimension's table or it could be due to using table from two different schema's
    what you say... i need your point of view
    regards,
    imran

  • What is detailed report and aggregated report

    Hi gurus,
    We generally tell cube gives the aggregated report and ODS gives detail report could any body tell me what exactly meaning behind there
    Please give me a good example for make me good under stand..
    Thanks in the advance
    Kiran
    Edited by: kiran kumar on Jan 16, 2008 10:34 AM
    Edited by: kiran kumar on Jan 16, 2008 10:35 AM

    Hi,
    aggreagted and detaliled are ways to mention how data is stored.
    when you stay data is summarized..it means...you have each and every record in source present in the same format as in BW.
    This is possible only if the data stored in the R/3 table is stord in BW in another table with the same keys or make sure the keys are in a way that you can identify each and evry record.
    This canbe dome through DSO as it has keys and you can make it load all the data in R/3 in exactly the same way.
    In case of cube you dont have any cubes and you have no control on aggregation of values also...so if 10 chars are same then key figues will add up.Generally cubes are specific with less number chars then DSO.
    if you want you can put lot of char and do reporting onto it.
    So if want each and very detail go for something whic stores data at the detailed level that is DSO.
    for aggregated so for cube since it aggrgated the data if the records repeat.
    I hope thats clear
    Thanks
    Ajeet

Maybe you are looking for