hive-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jamack, Peter" <PJam...@consilium1.com>
Subject Re: Medical App
Date Sat, 18 Feb 2012 19:58:06 GMT
There are ways to integrate each, but you'll have to figure out what
actually matches and what doesn't.  Maybe a customer id can link a medical
image, with their info, etc.
I've done a combo of MongoDB as the non structured OLTP, Postgres as the
RDBMS, and then 
Utilizing Hadoop as the analytic portion.  Maybe it was overkill but Hbase
wasn't that great of an option, so.  You might also need to use something
like Solr to index the non structured data.

 These days many ETL tools like Pentaho Kettle, Talend, even Informatica
and 
Spring Data-Hadoop have ways to integrate and connect.  I'm in the process
of figuring out the Pentaho Kettle big data integration and will follow up
with anything I learn in making it easier to integrate the different data
pieces.

Peter J

On 2/18/12 5:25 AM, "Dalia Sobhy" <dalia.mohsobhy@hotmail.com> wrote:

>Hiii all,
>
>I am developing a medical app but i have reached this:
>1. For structured Data such as patient info, medical history...etc: i
>would use RDBMS.
>2. For non-structured data such as scans, images, MRIs: i would use hbase
>"nosql database"
>
>So is this new arch is good??
>I need your opinions..
>If so how could i link between both sql and nosql, how could i link data..
>
>Sent from my iPhone


Mime
View raw message