I am currently deploying one Rasa instance in a productive environment. But how many concurrent users can be handled by a single Rasa instance?
Hi @matthiask this largely depends on the size of your model and some other factors. Can I ask how many concurrent users you expect with your bot?
I am expecting like 60 or 70 concurrent user. My model includes between 1500 and 2000 utterances. The server will have 4 GB RAM and 2 CPU cores.
that will be fine with one instance of rasa no problem not sure which version of rasa you’re on, but anything upward of 1.3.0 is best
How to decide on server configurations for rasa bot in production? Are there any parameters which we can use for identifying concurrent users?
Here, i can find minimum and recommended requirements. For a not complex model with around 5 happy path scenarios and their unhappy pathes and about 200 utterances. How many concurrent users should i expect it to be able to handle? When should i start thinking about moving to a cluster based deployment?