Skip to main content
January 16, 2024
Solved

Self-hosted Onestream in AWS

  • January 16, 2024
  • 4 replies
  • 0 views

Hi,

I didn't see anything in the Install/Config guide or in the Onestream Community search, so I figured I'd just ask.
Is there documentation regarding installing a self-hosted OneStream setup into a cloud environment?
Particularly regarding the registration/addition of new server instances (consolidation,staging, web, etc) into the config files:
-- Web Server Configuration file\Application Servers
-- Application Server Configuration file\Server Sets

Between server lifecycle events and lambdas calls, I'm sure something can be hacked together, but just thought I'd confirm what is/isn't available.

Thanks!

Best answer by JackLacava

The only "cloud features" you can have are related to Azure. There is nothing AWS-specific, as far as OneStream is concerned it will be just another Windows Server on-premises environment.

From your post, I guess you're looking for some sort of instance-scaling integration. At the moment, I'm not aware of any such feature available for on-premises setups. Happy hacking 👾

4 replies

Employee
January 16, 2024

I'm not an infra guy but isn't a self hosted cloud env the same as a self-hosted non cloud env except you don't own the hardware?

January 17, 2024

Agree with DanielWillis .  We use Google Compute Engine (Google Cloud-based VMs).  Same install & setup as on-prem.  You'll want to use FQDNs so the usual added DNS records for that is an extra step.  

January 17, 2024

The only "cloud features" you can have are related to Azure. There is nothing AWS-specific, as far as OneStream is concerned it will be just another Windows Server on-premises environment.

From your post, I guess you're looking for some sort of instance-scaling integration. At the moment, I'm not aware of any such feature available for on-premises setups. Happy hacking 👾

January 17, 2024

Yep, I was definitely referring to an Auto Scaling environment based off of dynamically set CPU & memory limits being surpassed to spin up a new instance, and then register that new server within the environment. It seems like it only becomes an issue during closing, at which point having a new instance spin up might help.

But no worries, you've all confirmed what we thought. A "hacking" we will go 🙂

Thanks much!