HBase summary tables

[Sarnath K <st...@gmail.com>]

Hi,

We are evaluating Apache Kylin for one of our requirements. We are
interested to know if we can perform aggregations through Storm and store
the output to HBase in a way that Kylin can believe it to be Summary table
of some Hive table. (We will also move raw data to hive).

Is this possible? We are trying to see if we can bypass MR cube builds. Any
update on this would be great!

Thanks,
Best,
Sarnath

Best,
Sarnath

Re: HBase summary tables

[Sarnath K <st...@gmail.com>]

Went through the streaming doc. I think it is a cool reuse of design by
using the job manager to integrate new realtime segments with traditional
ones. Nice one. HBase is a great way to invest such streaming aggregations.
Natural fit.
Thank you,
Best,
Sarnath

Re: HBase summary tables

[Sarnath K <st...@gmail.com>]

Yes, that's what I was looking for. I will try that out.
Best,
Sarnath

Re: HBase summary tables

[Luke Han <lu...@gmail.com>]

In that way, you may pull data from a streaming source like Kafka, right?

You could try the streaming cubing in 1.5.x

Thanks.

Luke


Best Regards!
---------------------

Luke Han

On Tue, May 10, 2016 at 9:29 PM, Sarnath K <st...@gmail.com> wrote:

> Hi,
>
> We are evaluating Apache Kylin for one of our requirements. We are
> interested to know if we can perform aggregations through Storm and store
> the output to HBase in a way that Kylin can believe it to be Summary table
> of some Hive table. (We will also move raw data to hive).
>
> Is this possible? We are trying to see if we can bypass MR cube builds. Any
> update on this would be great!
>
> Thanks,
> Best,
> Sarnath
>
> Best,
> Sarnath
>