You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kylin.apache.org by Hong Wei <hw...@qq.com> on 2016/07/12 09:42:31 UTC
my confusion with kylin detail query
Dear Kylin:
kylin 1.5.2
hbase 0.98
I am very confused in a little experiment for testing detail query with kylin. First, I import a table 'COFFEE' contains about 4258 records, and then build a simplest cube (only one dimension named 'MARKET', and one measure 'COUNT').
(1) When I query with SQL 'select count(*) from COFFEE', it shows '4258'.
(2) When I query detail with SQL 'select * from COFFEE', it shows only 4 records:
(3) When I query with SQL 'select market, count(*) from COFFEE group by market', it shows different results with the previous query:
(4) When I query with SQL "select * from COFFEE where market = 'Central' ", one ressult is shown.
How to explain and understand these results, and what should be noted when we build a cube expected to be able to query detail data ? Thank you.
回复： my confusion with kylin detail query
Posted by Hong Wei <hw...@qq.com>.
Hello Yiming Liu,
Thanks, I get it.
------------------ 原始邮件 ------------------
发件人: "Yiming Liu";<li...@gmail.com>;
发送时间: 2016年7月12日(星期二) 晚上6:33
收件人: "dev"<de...@kylin.apache.org>;
主题: Re: my confusion with kylin detail query
Hello Hong Wei,
Kylin is designed to support OLAP query. If you need query the detail data,
please enable the RAW measure when cube building.
More detail: http://kylin.apache.org/blog/2016/05/29/raw-measure-in-kylin/
2016-07-12 17:42 GMT+08:00 Hong Wei <hw...@qq.com>:
> Dear Kylin:
> kylin 1.5.2
> hbase 0.98
>
> I am very confused in a little experiment for testing detail query
> with kylin. First, I import a table 'COFFEE' contains about 4258 records,
> and then build a simplest cube (only one dimension named 'MARKET', and one
> measure 'COUNT').
> (1) When I query with SQL 'select count(*) from COFFEE', it shows
> '4258'.
> (2) When I query detail with SQL 'select * from COFFEE', it shows
> only 4 records:
> (3) When I query with SQL 'select market, count(*) from COFFEE group
> by market', it shows different results with the previous query:
> (4) When I query with SQL "select * from COFFEE where market =
> 'Central' ", one ressult is shown.
>
> How to explain and understand these results, and what should be
> noted when we build a cube expected to be able to query detail data ? Thank
> you.
>
>
>
--
With Warm regards
Yiming Liu (刘一鸣)
Re: my confusion with kylin detail query
Posted by Yiming Liu <li...@gmail.com>.
Hello Hong Wei,
Kylin is designed to support OLAP query. If you need query the detail data,
please enable the RAW measure when cube building.
More detail: http://kylin.apache.org/blog/2016/05/29/raw-measure-in-kylin/
2016-07-12 17:42 GMT+08:00 Hong Wei <hw...@qq.com>:
> Dear Kylin:
> kylin 1.5.2
> hbase 0.98
>
> I am very confused in a little experiment for testing detail query
> with kylin. First, I import a table 'COFFEE' contains about 4258 records,
> and then build a simplest cube (only one dimension named 'MARKET', and one
> measure 'COUNT').
> (1) When I query with SQL 'select count(*) from COFFEE', it shows
> '4258'.
> (2) When I query detail with SQL 'select * from COFFEE', it shows
> only 4 records:
> (3) When I query with SQL 'select market, count(*) from COFFEE group
> by market', it shows different results with the previous query:
> (4) When I query with SQL "select * from COFFEE where market =
> 'Central' ", one ressult is shown.
>
> How to explain and understand these results, and what should be
> noted when we build a cube expected to be able to query detail data ? Thank
> you.
>
>
>
--
With Warm regards
Yiming Liu (刘一鸣)