Re: how to use partitions once created?



What you really want to do is:

SELECT
{[Measures].[Amount],[Measures].[Sh Num]} ON COLUMNS,
{Crossjoin([Fund].[Fund Name].Members, { [Account].[State Abbr].[CT] } )}
ON ROWS
FROM [Investments]

i.e. specify which member you want in the set itself. The WHERE clause in
MDX is not meant for scenarios like this -- it is more for something like
this:

SELECT
{[Measures].[Amount],[Measures].[Sh Num]} ON COLUMNS,
{[Fund].[Fund Name].Members}
ON ROWS
FROM [Investments]
WHERE [Account].[State Abbr].[CT]

i.e you want all the funds but sliced for the CT state.

Btw, you typically specify a slice as an MDX expression when you create the
partition -- e.g. if you partitioned by states, it might be an expression:
[Account].[State Abbr].[CT]

Thanks,
Akshai
--
This posting is provided "AS IS" with no warranties, and confers no rights
Please do not send email directly to this alias. This alias is for newsgroup
purposes only.

"Rich" <Rich@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:788FD303-6476-4EC3-9EED-30BE4871881A@xxxxxxxxxxxxxxxx
> Thank you for your reply to my question. May I ask, how do you specify a
> slice ona partition? For example, my cube is called "Investments". One
> of
> the partitions is called "Conneticut"
>
> SELECT
> {[Measures].[Amount],[Measures].[Sh Num]} ON COLUMNS,
> {Crossjoin([Fund].[Fund Name].Members, [Account].[State Abbr].Members )}
> ON ROWS
> FROM [Investments]
> --Where
>
> This query yields something like (without any where clause):
>
> Amount Sh Num
> 20th-Century Select CA $30,000.00 1,111.10
> CT $38,000.00 1,421.62
> FL
> Acorn International CA
> CT $245,841.30
> FL
>
>
> I tried adding something to the Where Clause
>
> Where [Account].[State Abbr].CT
>
> but got this error message "Unable to open cellset
> Formula error - duplicate dimensions across (independent) axes - when
> calculating a query axis"
>
> Well, even though this is probably not a partition slice, how could I get
> this slice to work?
>
> Thanks,
> Rich
>
>
>
> "Akshai Mirchandani [MS]" wrote:
>
>> Partitions are physical objects -- they allow better management of the
>> data
>> in a cube and in some ways improve performance. But you don't (and can't)
>> query them directly -- the cube is the object that you care about.
>>
>> The "better management and performance" include:
>> - creating rolling partitions so that old partitions don't need to be
>> re-processed and new data can be incrementally added into the "current"
>> partition
>> - specifying slices on partitions so that partitions outside the slice in
>> the query don't need to scanned
>> - etc.
>>
>> Thanks,
>> Akshai
>> --
>> This posting is provided "AS IS" with no warranties, and confers no
>> rights
>> Please do not send email directly to this alias. This alias is for
>> newsgroup
>> purposes only.
>>
>> "Rich" <Rich@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
>> news:D8600CDE-4B62-4D37-BCA8-8C733EC5340A@xxxxxxxxxxxxxxxx
>> > Hello,
>> >
>> > I have a data warehouse/datamart database (please correct me what the
>> > correct name is for the database containing the cubes) containing 2
>> > cubes
>> > called "Dividends" and "Investments". I created 3 partitions in the
>> > "Investments" cube. I understand that partitions are a way to group
>> > data
>> > in
>> > a cube that is getting large. My question is how is a partition used
>> > once
>> > created (and processed and so forth)? the "Investments" cube has 2
>> > partitions "Conneticut", "Vermont".
>> >
>> > I opened up the MDX application and can only see the 2 cubes in this
>> > database for querying. Is it possible to query either of the
>> > partitions
>> > "Conneticut" or "Vermont"?
>> >
>> > Any explanations appreciated on the usage/purpose of cube partitions.
>> >
>> > Thanks,
>> > Rich
>> >
>> >
>>
>>
>>


.



Relevant Pages

  • Re: Multiple Cubes Sharing mdx Calculations?
    ... "I still see only 1 Fact table in the Cube browser. ... Different Fact Tables for Partitions ... view, from different data source views, or from different data sources. ... Cube partitions each have a slice for each attribute. ...
    (microsoft.public.sqlserver.olap)
  • Re: SSAS 2005 perform worse performance that 2000
    ... The slicer parameter is recommended but optionnal for your partitions ... recomend change memory settings. ... Our cube files takes about 35 GB, but database Estimated Size from cube ... queries from cache filtered 500000, Total queries from cache direct 30000. ...
    (microsoft.public.sqlserver.olap)
  • Re: Multiple Cubes Sharing mdx Calculations?
    ... structure as the fact table in the cube browser: ... Different Fact Tables for Partitions ... view, from different data source views, or from different data sources. ... Cube partitions each have a slice for each attribute. ...
    (microsoft.public.sqlserver.olap)
  • Re: Parallel process on 64Bit server with 4 CPUs
    ... The first partition to process in a cube with a private dimension will ... process any private dimensions and initialize the cube structure, ... before kicking off all the other "threads" to process the partitions. ... > We are using the Parallel process utility that shipped with the MS SQL ...
    (microsoft.public.sqlserver.olap)
  • Re: Cube does not show data when specifying aggregations
    ... I tried defining the percentage for 12 partitions at a time, ... When I tried for the first 12 partitions, cube had data. ... What I am doing is designing the aggregations for each partition, ... after migration, ...
    (microsoft.public.sqlserver.olap)