Fork me on GitHub

I’ve got an app which I would like to prototype with Datomic Cloud Ions. Though it is a chat not so I’m curious what the cold start/latency story is like. I’m having trouble finding information regarding this. Datomic Ions may not be suitable which is fine but if it is ok then I would prefer to use it.


The startup time is good. Clojure isn't run on the lambda, it runs in the auto scaling group, the lambda is a lightweight rpc proxy.


Could you explain what the rpc proxy is, like the work it is doing? I’m still learning all of this.


@ btw thanks for responding. Was thinking there is little activity in here and would go without a response


I shouldn't have said rpc. I made that bit up, I meant to say "it acts as an rpc". Basically you can think of it as doing a HTTP request to your auto-scaling group, forwarding on the data that went to the lambda to your asg.


I'm approximating a little bit here 🙂


chat not or chat bot?


It’s not a big deal. Once a conversation starts the lambda is warm, use one lambda for all bot calls and it will always be warm.