Does Memgraph have replication and how it works?

Question

I was checking Memgraph docs and wondering whether Memgraph has replication and why is that feature even useful? Is it for high availability? Does it allow queries to be parallelized? Something else?

Answer 1

In Memgraph, every read query is executed in its own thread, so you get the bonus from running queries in parallel with the respect to your number of cores.

The data is replicated in whole in Memgraph, with 2 replicas. So the engine can actually query either one of those machines. Memgraph is not yet distributed, i.e. it doesn’t shard the data, but keeps the same data in all the replicas. However, distributed Memgraph is currently in the works.