How to divide big RDF file to multiple small files & use embedding?

Users Dgraph
1

Suppose I have a big RDF file, called “A”. I divide “A” to small “B” & “C” files. The file “C” depends on file “B” . So The main question is: How to use file “B” in file “C” & import file “C” in Dgraph database? Is this possible at all? please help me.

2

You can use https://docs.dgraph.io/deploy#bulk-loader or https://docs.dgraph.io/deploy#live-loader
Both will use all RDFs in a given path. But this in a single bulk/batch. If you do loads on distinct moments they will not relate.

1 Like
3

Hi, How to create gzipped file from multiple RDF files? Should I just compress them to “.tar.gz” or “.zip” file? What about the relation between rdf files. How to make those files to relate to each other?

4

You don’t need to create any gzipped file. If you use “dot” in the -r flag Dgraph will seek for RDF, GZ or any other type compatible.

e.g:

./dgraph bulk -r . -s 1million.schema --zero localhost:5080 --map_shards=4 --reduce_shards=2

Just use the same Blank node in all RDFs - check this part of the tour A Tour of Dgraph

1 Like
5

Is “-r .” recursive? I mean does this flags makes bulk loader to search inner folders?

6

Yes, It will seek for any RDF below that path. Including inner folders.

1 Like
7

Thank you so much for your help.

1 Like