How is a gremlin query executed?

[Antriksh Shah]

Hey everyone,

Any pointers to where I can read more about how exactly a gremlin query gets executed?

I want to understand:

1) If there are pointed lookups like g.V().has(something).values()

 This will essentially be gremlin console making a call to graphindex(considering it is indexed), figure out the corresponding vertex partition of edgestore and return the data.

2) But when the queries are like g.V(1).repeat(both().where(criteria1)).until(criteria2), how does this get executed? 

I believe there would be a layer which will perform filtering which is in between the gremlin console and the backend? What is that layer? Or is it gremlin console gets the data for each repeat iteration, performs where clause then again makes a call to cassandra? 

Thanks and Regards,

Antriksh Shah

 

[Antriksh Shah]

Bumping this question up, if anybody has any suggestions.

[Sergey Volkov]

Did you try profile step? http://tinkerpop.apache.org/docs/current/reference/#profile-step