Bulk load a new graph using CQLOutputFormat

[cdegroc]

Hi 👋

I'm using the CQLInputFormat to dump data from JanusGraph using MapReduce/Spark.
I've noticed JanusGraph does not provide a CQLOutputFormat to do the opposite operation (load data).
I was wondering if the community had experience with Cassandra's CQLOutputFormat to bulk load a new graph.
Unless I am missing something, I guess a bulk load + a reindex could maybe be faster than using the SparkGraphComputer and OLAP traversals?

Any feedback appreciated. Thanks!

[li-boxuan]

I have no relevant experience but it sounds interesting! Maybe it is worth starting a discussion on the TinkerPop user mailing list too. I presume this is not specific to JanusGraph.