Re: hbase table creation

[Jean-Daniel Cryans <jd...@apache.org>]

(moving to the user ML)

in the shell (bin/hbase shell), type: help "create"

And then you will have the help for creating table along with examples.

Regarding the actual modeling of your table, putting the userid as the
row key (no need for a family) could be a viable option but the rest
seems weird. Maybe a better option would be to have a very tall table,
and followup comments would have a row key that would place them
directly next to the commented comment. Something like

eventid_comment1
eventid_comment1_comment1
eventid_comment1_comment2
eventid_comment1_comment2_comment1
eventid_comment2
eventid_comment2_comment1

This way a single scan can return all the comments regarding the event
that's originally being commented on, while enabling deep levels of
comments. Then you only need 1 family that describes all the
attributes of that comments (userid, whatnot).

I would try to use some sort of data in the row key component of a
comment, not numbering like in my example. Also watch out for the size
of those row keys, the bigger the more space every value will take.

J-D

On Tue, Feb 22, 2011 at 6:54 AM, Davison <sr...@nesote.com> wrote:
> i am new to hbase and hadoop. Any how i have succeeded in setting up a hadoop
> cluster which consists of 3 machines. Now i need some help on building up the
> database. I have a table "comments" contains fields(column families)
> 1)user id
> 2)comments
> 3)comments on comments(which can be more than one) and a status field to the
> same say,
>
> Could any one help me out to build the same using hbase/shell??
>
> Regards
>
>