Hi all,
is there any option to explicitly use data replication across stardog instances that are not part of the same cluster?
There's obviously a replication mechanism for nodes that are inside the same cluster. Wondering whether that could also be used to keep instances in different environments in sync.
Thanks,
Marcel
Like georeplication? I don’t think the current replication mechanism would work because it gives strong consistency guarantees so i would expect it to be very slow across high latency connections.
I think it would be possible to write a Kafka connect source and sink that used the Stardog transaction listener service and replicate that way. There might be some issues with blank nodes so you might have to set the preserve blank nodes option. That would also make it possible to to do some other cool stuff along with georeplication.
Thanks Zachary.
We're rather aiming at replication across environments, but
geographically close. Also, it's about smaller incremental updates. So
latency shouldn't be an issue.
Blank nodes are no issue either. Thus, the listener idea sounds
interesting, although it could be an overkill for our scenario. We'll
probably take a look, thanks for the suggestion!
Cheers,
Marcel
This would be a good place to start stardog-examples/examples/connectable at develop · stardog-union/stardog-examples · GitHub
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.