I'm wondering if this is a supported scenario in ElasticSearch, reading the
guide and API reference I couldn't find a way to achieve this.
I'd like to index documents only once, say in a master index, and then
create secondary or "meta" indices that would only contain a subset of the
master index.
For example, document A, B and C would be indexed once in the master index.
Then a secondary index would be able to see only documents A and B, while
another secondary index could see only documents B and C, etc. (and by
"see" I mean the search queries should only consider those documents)
The idea being that documents could be relatively big, and they should not
be indexed multiple times.
Does that make sense? Am I missing "the right way" to design such a
pattern? I am new to ES.
I'm wondering if this is a supported scenario in Elasticsearch, reading
the guide and API reference I couldn't find a way to achieve this.
I'd like to index documents only once, say in a master index, and then
create secondary or "meta" indices that would only contain a subset of the
master index.
For example, document A, B and C would be indexed once in the master
index. Then a secondary index would be able to see only documents A and B,
while another secondary index could see only documents B and C, etc. (and
by "see" I mean the search queries should only consider those documents)
The idea being that documents could be relatively big, and they should not
be indexed multiple times.
Does that make sense? Am I missing "the right way" to design such a
pattern? I am new to ES.
On Monday, June 16, 2014 3:51:45 AM UTC+2, Martin Angers wrote:
Hi,
I'm wondering if this is a supported scenario in Elasticsearch, reading
the guide and API reference I couldn't find a way to achieve this.
I'd like to index documents only once, say in a master index, and then
create secondary or "meta" indices that would only contain a subset of the
master index.
For example, document A, B and C would be indexed once in the master
index. Then a secondary index would be able to see only documents A and B,
while another secondary index could see only documents B and C, etc. (and
by "see" I mean the search queries should only consider those documents)
The idea being that documents could be relatively big, and they should not
be indexed multiple times.
Does that make sense? Am I missing "the right way" to design such a
pattern? I am new to ES.
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.