Date
1 - 2 of 2
MixedIndex naming convention
Ravikumar Govindarajan <ravikumar....@...>
I saw in many places of documentation/tutorials, that the mixed indexes have this mgmt.buildIndex("vertices", Vertex.class).addKey(key).buildMixedIndex(INDEX_NAME); // With INDEX_NAME ='search', mostly Will this create one index in ElasticSearch/SOLR for one property or all properties are clubbed under a single index? Also does JanusGraph partition these text based indexes, in case they get too large? -- Ravi |
|
Jason Plurad <plu...@...>
I think all of the answers are already in the docs (see the Note box in the ES Configuration Overview and Index Creation Options). If there are specific ways you think the docs could be improved, it would be good if you opened up an issue and even better if you submitted a pull request. In your graph configuration, you actually define the indexes and the shards. You can define more than one index by using a different name for [X]. If you don't set the number of shards, it will default to 5 as dictated by Elasticsearch. For example:
After you initialize your graph, you can verify that 2 indexes are created in ES, with different # shards in this example:
-- Jason On Tuesday, June 20, 2017 at 8:31:56 AM UTC-4, Ravikumar Govindarajan wrote:
|
|