Hi Manoj,

Usually, we know the query pattern first, and then design the cube
according to it.

For your question, if the query is among the 10 normal dimensions, the
query should be fast; If you have derived column in query, Kylin need do a
mapping from the FK to derived columns; If the derived column's cardinality
is much less than FK's, that means a lot of aggregation would happen at
query time, the performance would be impacted.


2017-12-18 12:19 GMT+08:00 Kumar, Manoj H <[email protected]>:

> Can you pls. advise on this case – I have Cube built on for below case –
> It has 234 dimension(224 dimensions are for Derived & 10 dimensions from
> Fact table which are normal one).
>
>
>
> Fact table – 1
>
> Dimension/look up table – 9 [ All these are linked to Fact table column
> Mapping]
>
>
> ------------------------------
>
> *Model Name*
>
> Data_model
>
> *Cube Name*
>
> CUBE_1
>
> *Fact Table*
>
> V_EP_FACT_ACTUALS
>
> *Lookup Table*
>
> 9
>
> *Dimensions*
>
> 234
>
> *Measures*
>
> 13 (Sum of Columns)
>
>
>
>
>
> User will be using Tableau Desktop to get the data viewed – which way I
> should write the query to get faster response from Hbase Cube which
> includes Dimensions & Fact table. Pls. advise. Pls. let me know if you need
> any other information.
>
>
>
> Regards,
>
> Manoj
>
>
>
> This message is confidential and subject to terms at: http://
> www.jpmorgan.com/emaildisclaimer including on confidentiality, legal
> privilege, viruses and monitoring of electronic messages. If you are not
> the intended recipient, please delete this message and notify the sender
> immediately. Any unauthorized use is strictly prohibited.
>



-- 
Best regards,

Shaofeng Shi 史少锋

Reply via email to