Resources management?
Hello, I'm new to Railway.
I currently have the Developer Plan.
I was wondering if there is a way to manage the resources available to a single project/instance, I would like to upload multiple different projects in my account and there are projects that might need more vCPU than the others or RAM.
My current plan says: "8 GB of RAM, 8 vCPU, and 100 GB of Shared Disk"
Are those equally shared between all services? is there a way to configure something like percentages to each project or something like that?
I want to know if there is currently any way to configure the available resources, thank you.
19 Replies
Project ID:
N/Anope, its 8 gigs and 8cpu per project :)
Oh very nice!
:)
as for configuring the resources, thats something i dont know, maybe we should wait for a conductor
Thank you, really, now I was wondering, how about the pricing?
In the same scenario, I have a project more resource consuming than the other, both of them will run instances with 8cpu and 8 gigs? wouldn't that be costly? or it scale up automatically? how about that?
you only pay for what you use
so if you have 8 gigs but use 100 megs
you only pay for 100 megs
its very nice tbh
Thank yoouuu
np my dude :)
I'm guessing that if I spin up a Node app it will only be using 1 vCPU per instance since it's single threaded so if I wanted I could scale it up to 8 instances in the same project, right?, that's kind of the way my thoughts are getting, it is something like that?
🥹
no, i think
you can have replicas, but i dont know how those work
All right then, thank you very much for all 🥹
np man, glad i was able to help
fun fact you can actually use 0.0 vCPU, if you write efficient code, don't do heavy computational tasks and don't have much traffic (if it's a web app) your costs can be very cheap
yeah, all my stuff uses 0.0
87 cents, that's expensive
ram
and yes
i know go would use like 5 megs
:)
If I ever need some heave computational task, is it recommended to go for a lamda instead? or railway is already cheap enough?
if your background resource usage is low, and you do a few quick computationally heavy tasks a day, your average usage will still be low, so low costs too, but of course do your own cost analysis