Multithread implementation

[helviojunior]

I have implemented a multithread version, for me it was at least 10x faster.

To do that i splited the queryes read only to write, to reduce deadlocks.

[arvchristos]

Hey @helviojunior !

I tried importing a fairly large dataset with your approach and I find it less performant than the existing approach (disclaimer: I optimized the current approach and introduced async). This may be because of my dataset. Do you have any statistics on the size of the dataset you used?

[enj5oy]

Hello @arvchristos !
Is it typical for the upload of 750 computer objects to take around 100 seconds?

[arvchristos]

Hey @enj5oy ,
My experience is that it highly depends on the connected nodes of the computer objects. Do you have more statistics on the file size? (750 computer objects are not that much but maybe the created Neo4j entries are far more).

[enj5oy]

@arvchristos upload of 750 computer objects takes place within 100 seconds. The file with all 7500 computer objects occupies 68 MB