We do deployments and customize things for users. When we deploy PredictionIO
we typically have one machine that is for only PIO permanent servers. It runs
the PredictionServer (started with `pio deploy`) and the EventServer (started
with `pio eventserver`). These services communicate with
Dear Pat,
Thanks for the detailed guide. It is nice to know it is possible.
But I am not sure if I understand it correctly, so could you please
point out any misunderstanding in the following? (If there is any)
Let's say I have 3 machines.
There is a machine [EventServer and data store)
Yes, this is the recommended config (Postgres is not, but later). Spark is only
needed during training but the `pio train` process creates drives and executors
in Spark. The driver will be the `pio train` machine so you must install pio on
it. You should have 2 Spark machines at least because
