I inherited a project that requires efficient data handling and query execution. I'm interested in optimizing performance and I'd like to know about the feasibility of parallelizing query execution using the Memgraph C++ client.
Is it possible to execute multiple queries in parallel directly through the Memgraph C++ client? If it can be done, what is the recommended approach to implement this? Are there specific patterns to follow (e.g., managing connections, handling concurrency, etc.)?
The mgclient doesn't expose specific parallelization primitives (every instantiated client/connection is separated, independent, and synchronous). Despite that i t's possible to execute queries in parallel by instantiating the client (example can be found at basic.cpp file) inside multiple C++ threads. Basic parallelization should be possible in a reasonably simple way.