4

Imagine I have graph data that is beyond the size of a single machine.

How would you shard a graph database?

I asked on Hacker News and people suggested sharding based on a hash of the predicate-subject-object, ala RDF triples.

I could partition the graph across replicas and query all replicas to aggregate answers.

But what about joins across machines?

Do I need to colocate join keys on all replicas?

Or do I just send the query to all replicas and in parallel and aggregate the responses?

Improve this question
3
  • You haven't mentioned a database vendor, but I imagine sharding should be mostly transparent to you. Joining, querying — all those operations should be handled by the database so your code doesn't need to. Can you edit your question to include more information about the database vendor you are using? Bear in mind that questions about how to configure a database are likely off-topic. Tool questions belong on StackOverflow or another relevant community.
    – Greg Burghardt
    Commented May 25, 2023 at 11:21
  • I am the database vendor, in other words I am experimenting with database internals and I am curious about the implementation techniques of sharding graph data.
    – Samuel Squire
    Commented May 26, 2023 at 11:17
  • I see. This is more a database architecture question. I see two close-votes for "needing focus". Don't let multiple question marks fool you. You cannot properly shard data unless you consider the other questions in this post.
    – Greg Burghardt
    Commented May 26, 2023 at 12:11

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.