Is there any intention for prisma to natively support having separate read and write clients? For context, we (wearseasons.com) are working to scale our DB. One step of that process is moving to a DB cluster with one "write" node and a set of "read" nodes which are all downstream of the leader "write" node. To do this we need to send mutations to the write node and read requests to the load balancer sitting in front of the read nodes. Since this is a pretty common thing for eng teams to do, I was curious if prisma has any plans (or existing functionality?) to take in separate read/write DB URLs and automatically route things accordingly. cc @Luc Succes

For added context, here’s a github issue that describes the problem more in depth. https://github.com/prisma/prisma/issues/172

@Faiyam Rahman 👋 You can implement it in this way 🙂

Thanks. That's actually what we're doing now. We created a wrapper around a read client and write client that routes a given query to a specific client based on whether it's a read or write. We were hoping there might be some talk of natively supporting this functionality. Seems like that's still in the works