hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ding, Hui" <hui.d...@sap.com>
Subject RE: More examples for using hbase
Date Wed, 10 Sep 2008 19:43:00 GMT
Hi Stack,

Thanks for your response!

What do you mean by unit tests have lots of exercise? Can you be a
little bit more specific?
I will look at the REST package, thanks!

Right now I just want to explore whether and how hbase can accommodate
the traditional relational query operators in an efficient/scalabe way.
For example, can I easily retrieve the rows whose column A has a value
equal to/greater than something? Or should I iterate through all rows
and check?

Also, I guess hbase+mapreduce is not good for performing joins, so how
should one organize the data to avoid performing joins? More
specifically, say I have
lots of movie info, each contains a number of stars that are in the
movie.  Should I repeatedly store every star's info for every movie row?
Or should I have another table just for the stars? Maybe I haven't
jumped out of the RDBMS model to fully realize the power of
hbase/bigtable...

Thanks a lot!
Andrew

-----Original Message-----
From: stack [mailto:stack@duboce.net] 
Sent: Wednesday, September 10, 2008 11:32 AM
To: hbase-user@hadoop.apache.org
Subject: Re: More examples for using hbase

Ding, Hui wrote:
> Hi all,
>
> I am a newbie to hbase, and I am just wondering where can one find
more
> examples of using hbase, other than the simple one given in the
"Getting
> Started"?
> I have read the bigtable paper, hbase architechture doc etc.
>   
Unit tests have lots of exercise of the hbase API.  That might be of
help?

Otherwise, you could look at REST/thrift implementations under the 
respective packages for how they grapple the hbase API.

What you need?  Maybe we can do some sketchs for you here up on this
list.

St.Ack



Mime
View raw message