Can I temporarily pause the RDF bulk loading process for a while and then recover it back to process?

Question

I am currently bulk loading DBpedia and Freebase data dumps into the virtuoso open source edition v7 using the rdf_loader_run() procedure. The database is on our server and I have another graph IRI in Virtuoso running for an online service.The loading process largely effect the responsing rate of the service. Can I use the rdf_loader_stop() in isql to temporarily pause the loading threads and is it capable to recover back to the former status of loading process by running rdf_loader_run() command again?
The Official Wiki doesn't explain this well.

Answer 1

rdf_loader_stop() will stop importing after files which are imported right now are finished. The queue is actually left intact at this point. Strictly speaking, it just puts a flag, which import-processes check before proceeding to the next file in queue.

rdf_loader_run() will continue with the next files in queue

Answer 2

rdf_load_stop() will stop importing, and you can restart it at any time by running rdf_loader_run() again.

If you need to brute force a stop, type txn_killall(1).

You can check the status of your import by opening a new isql window and type the followng query: SELECT ll_state, COUNT(ll_state) FROM load_list GROUP BY ll_state

ll_state = 0: files not yet processed
ll_state = 1: files that are currently being processed
ll_state = 2: files that are completed

If you did a brute force stop, you'll have to change the ll_state=1 rows to ll_state=0 again to make sure they will be processed.

Answer 3

rdf_load_stop() is what you are looking for. However, it just stops the next file from being loaded. The current file has to finish first.

It basically inserts a ##stop file in the db.dba.load_list table, which (I guess) gets checked first before Virtuoso processes the next file in the load_list