3.1. Requirements#
Note
This section is the same as the node requirements section - their requirements are very similar.
The (minimal) requirements of the node and server are similar. Note that these are recommendations: it may also work on other hardware, operating systems, versions of Python etc. (but they are not tested as much).
Hardware
x86 CPU architecture + virtualization enabled
1 GB memory
50GB+ storage
Stable and fast (1 Mbps+ internet connection)
Public IP address
Software
Operating system: - Ubuntu 18.04+ - MacOS Big Sur+ (only for node) - Windows 10+ (only for node)
Python
Docker
Note
For the server, Ubuntu is highly recommended. It is possible to run a development server (using vserver start) on Windows or MacOS, but for production purposes we recommend using Ubuntu.
Warning
The hardware requirements of the node also depend on the algorithms that the node will run. For example, you need much less compute power for a descriptive statistical algorithm than for a machine learning model.
3.1.1. Python#
Installation of any of the vantage6 packages requires Python 3.10. For installation instructions, see python.org, anaconda.com or use the package manager native to your OS and/or distribution.
Note
We recommend you install vantage6 in a new, clean Python (Conda) environment.
Higher versions of Python (3.11+) will most likely also work, as might lower versions (3.8 or 3.9). However, we develop and test vantage6 on version 3.10, so that is the safest choice.
Warning
Note that Python 3.10 is only used in vantage6 versions 3.8.0 and higher. In lower versions, Python 3.7 is required.
3.1.2. Docker#
Docker facilitates encapsulation of applications and their dependencies in packages that can be easily distributed to diverse systems. Algorithms are stored in Docker images which nodes can download and execute. Besides the algorithms, both the node and server are also running from a docker container themselves.
Please refer to this page
on how to install Docker. To verify that Docker is installed and running
you can run the hello-world
example from Docker.
docker run hello-world
Warning
Note that for Linux, some post-installation
steps may
be required. Vantage6 needs to be able to run docker without sudo
,
and these steps ensure just that.
For Windows, if you are using Docker Desktop, it may be preferable to limit the amount of memory Docker can use - in some cases it may otherwise consume much memory and slow down the system. This may be achieved as described here.
Note
Always make sure that Docker is running while using vantage6!
We recommend to always use the latest version of Docker.