Row with many columns

[ruslan usifov <ru...@gmail.com>]

Hello

For example if we have table, which have rows with many columns (10000 or
more) how this data will by partitioned?? Does row will by split by some
regions servers?

Re: Row with many columns

[ruslan usifov <ru...@gmail.com>]

Thank for answers. This moment is misunderstood for me. So i post this
message to both mailing list. Part this provoke by lack of documentation in
both products, and my language barrier :-))


Thanks

2010/2/21 Erik Holstad <er...@gmail.com>

> Hey Rusian!
> Both HBase and Cassandra are as you now know row partitioned, so to get the
> most out
> of any of these I think you should try to change the structure of your
> data,
> so that you for
> example have one object per row, like Ted suggested or in some other way.
>
>
> --
> Regards Erik
>

Re: Row with many columns

[Erik Holstad <er...@gmail.com>]

Hey Rusian!
Both HBase and Cassandra are as you now know row partitioned, so to get the
most out
of any of these I think you should try to change the structure of your data,
so that you for
example have one object per row, like Ted suggested or in some other way.


-- 
Regards Erik

Re: Row with many columns

[Ryan Rawson <ry...@gmail.com>]

Hi,

No, each row must be owned by a single regionserver... its the basis of our
consistency. However, you can make region size nearly arbirarily large.

On Feb 20, 2010 11:53 AM, "ruslan usifov" <ru...@gmail.com> wrote:

Hello

For example if we have table, which have rows with many columns (10000 or
more) how this data will by partitioned?? Does row will by split by some
regions servers?