Question 1:
I am trying to find some performance data or internal janusgraph documentation regarding how the performance of different operations are affected by the size of the data.
get/add/UpdateVertex(vertexName)
vertexName is the key and its used to create index.
Are these operations guaranteed to be constant time operations especially when index key is already specified in the operation?
Or does it depend on the number of vertices present in the graph and can take longer time for these operations as the data size grows?
Question 2:
JanusGraphFactory.open()
In one of my testing I closed the graph using JansGraphFactory.close() and immediately tried to open it using open() and it took 2 mins. This was clustered environment with Casandra backend.
Most of the other times same code behaved very fast and returned almost instantly.
Does the time it takes to open() graph depend on the data size?
Question 3:
Indexing/Re-indexing
As and when the data size grows with more vertices and edges in the graph, does Janusgraph automatically optimize or re-index the data or should applications try to re-index it explicitly every now and then?
I am sorry if these questions sound naive as I am a beginner here :)