atlas-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Christian Rieck (JIRA)" <j...@apache.org>
Subject [jira] [Created] (ATLAS-1294) Return Titan-id in Entity retrieval for faster gremlin queries
Date Tue, 15 Nov 2016 13:50:58 GMT
Christian Rieck created ATLAS-1294:
--------------------------------------

             Summary: Return Titan-id in Entity retrieval for faster gremlin queries
                 Key: ATLAS-1294
                 URL: https://issues.apache.org/jira/browse/ATLAS-1294
             Project: Atlas
          Issue Type: Improvement
            Reporter: Christian Rieck


When I do a discovery/search/gremlin?query I can either specify my starting node with the
titan id , query=g.v(id).... or if I don't have it I can use g.V.has('Referenceable.qualifiedName','id')....
The first takes about 200ms and the second takes roughly two seconds (timed in chrome dev
tools) . As such, I would like to get the titan id when I fetch an entity so that later queries
takes a reasonable amount of time. 

Alternatively, the search documentation must be updated with loop-query examples so I can
use the DSL instead. The technical user guide doesn't even mention Gremlin, so I am not sure
if it is supposed to be used or not. Since the web has many good Gremlin tutorials and no
Atlas DSL tutorials I would prefer to use Gremlin. 

As a use case, I have one time series entity that is updated daily with new values from csv-files.
The lineage shows n CSV files, n import_job entities and one timeseries, and i would like
to know which CSV-file is the source of a range of my timeseries. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message