#Help about Infra

5 messages · Page 1 of 1 (latest)

gusty topaz
#

Hi guys. I need help with two things:
1 - I need build online penny auction system (500 online users). I like Railway and want use because it is nice and i don't care about infra details. What is the best way to organize the infra in Railway for my django app for these 500 online users. Users can make bids on each second in my penny auction system. Is best have a mysql locally in railway and system in railway too or use supabase as database and app in railway?

2 - If i need more than one server, how sessions will work in django app in railway for multiple servers?

3 - There is a way to schedule database backup to AWS S3? How it works?

timber meadowBOT
#

Project ID: N/A

gusty topaz
#

N/a

verbal hazel
#

hello!

  1. most of what I would say is explained here - https://docs.railway.com/overview/best-practices

but I would like to put emphasis on keeping everything in the same project, meaning do not use supabase as latency to supabase can be detrimental to high throughput applications, instead use a database hosted on Railway in the same project.

  1. I strongly recommend using an async worker for gunicorn like uvicorn, then you can likely do 8 threads, and you can also add replicas to horizontally scale - https://docs.railway.com/reference/scaling#horizontal-scaling-with-replicas

  2. you can do backups to S3 on a schedule, but we do not provide a platform native way to do that, instead I recommend you check out our native backups solution - https://docs.railway.com/reference/backups

hope this helps!

verbal hazel
#

!s