I use it a lot for production ETL pipes, and sqlite3 is wonderful. The only issue we see is the writes slow down when the table rows hit 8–9 figures during a fr

I use it a lot for production ETL pipes, and sqlite3 is wonderful. The only issue we see is the writes slow down when the table rows hit 8–9 figures during a fresh load. In client/server, our biggest issue is data replication and sync. We have this application that use sqlite3 as a db for mostly read only web application. It really act as a cache server. We can’t figure out a clean way to update the data from remote systems. We end up writing a web socket process on the box to pull data from remote systems and write the data one batch at a time. This was/is a big pain point for us. Especially when updates are coming in fast.