hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mohammad Tariq <donta...@gmail.com>
Subject Re: HBase and Datawarehouse
Date Mon, 29 Apr 2013 17:35:37 GMT
Sorry for the late response. I totally agree with Anil. If you have
warehousing needs, I would also suggest Hive. You could easily map your
Hbase tables to your Hive tables and crunch crunch the data. It would save
you from writing lengthy and tedious MR jobs. And as Anil has said Pig is
another good choice, if you have to do lot of transformations on your data.

Warm Regards,
Tariq
https://mtariq.jux.com/
cloudfront.blogspot.com


On Mon, Apr 29, 2013 at 10:30 PM, anil gupta <anilgupta84@gmail.com> wrote:

> Inline.
>
> On Sun, Apr 28, 2013 at 10:40 PM, Kiran <kiranvk2011@gmail.com> wrote:
>
> > Anil,
> >
> > So it means HBase can help in easy retrieval and insertions on large
> > volumes
> > of data but it lacks the power to analyse and summarize the data?
>
> Out of the box, it can do simple aggregations like sum, avg, etc. But, if
> you have complex analytical queries(lead, lag rolling aggregates) then you
> can write your Coprocessor for doing those analytical queries.
>
> > In HBase
> > can't we write Map-Reduce jobs that can do this "data cunching"?
>
> Yes you can do. But, if you are only going to do MR then why use HBase for
> storing data?
>
> > As per your
> > analysis isn't that a feasible approach than the data warehousing
> systems?
> >
> I dont know your use case in detail so cant say whether it will work for
> you or not. But, theoretically it is feasible. Have you evaluated hive/pig
> for data warehousing?
>
> >
> >
> >
> > --
> > View this message in context:
> >
> http://apache-hbase.679495.n3.nabble.com/HBase-and-Datawarehouse-tp4043172p4043220.html
> > Sent from the HBase User mailing list archive at Nabble.com.
> >
>
>
>
> --
> Thanks & Regards,
> Anil Gupta
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message