Hi Parkavi, first of all, you need have a clear concept of dimension
columns and measure columns; There are many publications on this so I will
not elaborate; To be brief, the dimension columns are the columns that will
group by; The measure columns are the columns that will be aggregated
(COUNT, SUM, MAX, MIN, etc);

>From your description, Col1, Col2, Col3 and Col4 should be dimension
columns; and Col5, Col6 and Col7 are meansure columns; In the cube you can
pick Col1 to Col4 as dimensions, and define 3 measures: SUM(Col5),
SUM(Col6) and SUM(Col7);

After cube be built, you can run query:
select col1, col2, col3, col4, SUM(col5), SUM(col6), SUM(col7) from
facttable group by col1, col2, col3, col4;

To get the SUM of col5, col6, col7, you can use sub query, like:

select col1, col2, col3, col4, SUM(SUM5, SUM7) as SUM57, SUM(SUM5, SUM6,
SUM7) as SUM567 from (
  select col1, col2, col3, col4, SUM(col5) as SUM5, SUM(col6) as SUM6,
SUM(col7) as SUM7 from facttable group by col1, col2, col3, col4
) as temp;

In your case I didn't see a hierachy structure, so just use "Normal" for
the dimensions;

Kylin doesn't forward query to Hive so far; In the future it may integrate
with SparkSQL;

2015-05-13 18:27 GMT+08:00 Parkavi Nandagopal <[email protected]>:

> Hi,
>
> Please explain how to create cube for the below scenario:
>
> Fact table:
>
> Col1     Col2     Col3     Col4     Col5     Col6     Col7     Value
> L1            10           2              3              Tt1          Tp1
>        Pm1       23.2
> L1            10           2              3              Tt1          Tp1
>        Pm2       56.2
> L1            10           2              3              Tt1          Tp2
>        Pm1       12.7
> L1            10           2              4              Tt2          Tp1
>        Pm2       22.3
> L2            10           3              1              Tt1          Tp3
>        Pm3       11.6
> L2            11           3              1              Tt2          Tp1
>        Pm4       12.3
>
>
> I want like below combination:
>
> Col1, Col2, Col3, Col4,     Col5 + Col6 + Col7 = value
> Col1, Col2, Col3, Col4,     Col5 + Col6 = value
> Col1, Col2, Col3, Col4,     Col6 + Col7 = value
> Col1, Col2, Col3, Col4,     Col5 + Col7 = value
> Col1, Col2, Col3, Col4,     Col5 = value
> Col1, Col2, Col3, Col4,     Col6 = value
> Col1, Col2, Col3, Col4,     Col7 = value
>
>
> I tried with hierarchy dimension with Col1, Col2, Col3, Col4, Col5, Col6 ,
> Col7.
> It is creating with 7 dimensional cube, but I need only 3 dimension with
> Col5, Col6 , Col7.
>
> If I didn't select Col1, Col2, Col3, Col4 when I creating dimension, then
> I can't able to query Col1, Col2, Col3, Col4 details in query engine.
> Is Kylin 0.7.1 HOLAP?
> i.e data which is not in cube will redirect it to hive table?
>
>
>
> Thanks,
> Parkavi.
>
>
>
> ::DISCLAIMER::
>
> ----------------------------------------------------------------------------------------------------------------------------------------------------
>
> The contents of this e-mail and any attachment(s) are confidential and
> intended for the named recipient(s) only.
> E-mail transmission is not guaranteed to be secure or error-free as
> information could be intercepted, corrupted,
> lost, destroyed, arrive late or incomplete, or may contain viruses in
> transmission. The e mail and its contents
> (with or without referred errors) shall therefore not attach any liability
> on the originator or HCL or its affiliates.
> Views or opinions, if any, presented in this email are solely those of the
> author and may not necessarily reflect the
> views or opinions of HCL or its affiliates. Any form of reproduction,
> dissemination, copying, disclosure, modification,
> distribution and / or publication of this message without the prior
> written consent of authorized representative of
> HCL is strictly prohibited. If you have received this email in error
> please delete it and notify the sender immediately.
> Before opening any email and/or attachments, please check them for viruses
> and other defects.
>
>
> ----------------------------------------------------------------------------------------------------------------------------------------------------
>

Reply via email to