Install Server
Upsonic Server consists of two subsystems: the Dashboard and the API. Both systems run in a single container and connect internally. This document guides you through the installation and execution of the container without any issues.
To get started, you will need:
- The latest version of Docker installed and functioning properly
- TCP ports 7341 and 7340 ready for use
Deployment
First, we will create a volume to store everything:
docker volume create upsonic_on_prem_data
Setting .env
nano .env
admin_key=YOUR_ADMIN_KEY_FOR_API # 123
admin_pass=YOUR_ADMIN_PASS_FOR_DASHBOARD # 123456
admin_username=YOUR_ADMIN_USERNAME #upsonic_admin
after adjusting the settings, you can exit by pressing Ctrl+X, then confirming with ‘Y’ and pressing Enter.
Now, let’s proceed with the download, installation, and execution of the Upsonic Server container:
docker run --env-file .env -d -p 7341:7341 -p 7340:7340 --name upsonic_on_prem --restart=always -v upsonic_on_prem_data:/db upsonic/on-prem:v0.30.7-AMD64
After that, you should see:
root@server:~# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
918a4c74185a upsonic/on-prem "bash /run.sh" 11 seconds ago Up 10 seconds 0.0.0.0:7340-7341->7340-7341/tcp, :::7340-7341->7340-7341/tcp upsonic_on_prem
Suggested Version
As you can observe, we are recommending a specific version, v0.30.7
. This recommendation is based on our functionality testing and stability analysis.
If you wish to explore other versions, please visit the Release Notes page.
Extra Options
Limiting Model CPU usage
Limiting model CPU usage is improves the general performance of application. Because if you dont specify anything the model can take whole CPU to making analyses. But with this env variable you can limit how many CPU core that you want to give to model:
main_model_cpu=3
Logging In
Now, you can access your Upsonic Server via localhost or any other method of accessing your server:
https://localhost:7341