Hello,
A partner gave me editing rights to a graph from his Virtuoso triple store. However, he only gave me access to the HTTP sparql-auth interface and I would like to upload a 30 MB (281.000 triplets). I guess that launching a query with 281.000 UPDATE from the HTTP interface could cause a timeout. Is there another way to upload those triplets via sparql-auth ?
Thank you,
Certainly inserting 281K triples via the HTTP sparql-auth endpoint it not recommended, as that a single transaction performed in memory and would problem timeout/hang the system/OS. Such large datasets should really be loaded using the Virtuoso RDF Bulk Loader which was designed for loaded large dataset files.
To upload via the sparql-auth endpoint any insert queries should not be more than about 5K in size generally, which would also be dependant on the configuration of the Virtuoso instance and if it has been performance tuned for loading large datasets.