Hi,
I have recently started playing with ES for a site search project. My
search query runs against multiple tables across two databases, masterDb &
clientDb. So my plan is to create two indexes one for each database and
then add multiple types to each index.
And then finally query across those indexes for result.
Can anyone comment is this approach correct? And secondly how can i query
across indexes & types that I have created to get my results back? Any link
to a tutorial would be great
thanks
j
as to how to organize your data, it is really hard to say
it all depends of the size of your databases and the type of documents you
want to index, from the top of my head:
if your documents are hierarchically related (parent-child) then they
have to go to this same index
if you run your queries only against subset of tables (different query
different subset), then consider grouping them into different indexes (or
use routing)
if you run your query always against all tables, then you can even put
everything in one index with many shards (as there is really no much
difference between running query on one index with multiple shards and
running query on multiple indexes)
check this articles as starting point to learn more:
On Friday, December 27, 2013 10:22:04 PM UTC, jsp wrote:
Hi,
I have recently started playing with ES for a site search project. My
search query runs against multiple tables across two databases, masterDb &
clientDb. So my plan is to create two indexes one for each database and
then add multiple types to each index.
And then finally query across those indexes for result.
Can anyone comment is this approach correct? And secondly how can i query
across indexes & types that I have created to get my results back? Any link
to a tutorial would be great
thanks
j
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.