edge 12 min read

Edge System Requirements

Hardware, software, and network requirements for running Pyvorin Edge on ARM64 gateways.

Published Jun 2, 2026

Introduction

Before you install Pyvorin Edge, you need to make sure your gateway hardware and software environment can support it. This article covers every requirement in detail: the computers we support, the operating systems we test against, the Python version you need, memory and disk expectations, and network prerequisites.

Pyvorin Edge is designed to be lean, but it does perform compilation, SQLite storage, and optional cloud sync. A £35 Raspberry Pi Zero 2 W might struggle with compilation tasks, while a Raspberry Pi 5 will handle hundreds of sensors with ease. This guide will help you choose the right hardware.

Supported Hardware

Pyvorin Edge targets ARM64 (also known as AArch64) devices. This is the 64-bit instruction set used by modern Raspberry Pi boards, NVIDIA Jetson modules, and many industrial gateways. We do not support 32-bit ARM (armv7l) because the compiler bridge and NEON optimisations require 64-bit registers.

Recommended Devices

Device CPU RAM Status Notes
Raspberry Pi 5 BCM2712, 4x Cortex-A76 @ 2.4 GHz 4 GB or 8 GB Recommended Best performance, active cooling advised for sustained compilation
Raspberry Pi 4 BCM2711, 4x Cortex-A72 @ 1.8 GHz 2 GB, 4 GB, or 8 GB Recommended Most widely tested platform; 2 GB works for small deployments
Raspberry Pi 400 BCM2711, 4x Cortex-A72 @ 1.8 GHz 4 GB Supported Built-in keyboard form factor; identical performance to Pi 4
Raspberry Pi CM4 BCM2711, 4x Cortex-A72 @ 1.5 GHz 1 GB to 8 GB Supported Industrial module; ensure adequate heat sinking
NVIDIA Jetson Nano 4x Cortex-A57 @ 1.43 GHz 4 GB Community Works but compilation is slower; GPU not utilised by Pyvorin Edge
Orange Pi 5 Rockchip RK3588S, 8-core 4 GB to 16 GB Community Fast CPU; community-tested only

Minimum vs Recommended Hardware

Spec Minimum Recommended
CPU 4x ARM Cortex-A72 @ 1.5 GHz 4x ARM Cortex-A76 @ 2.4 GHz
RAM 2 GB 4 GB or more
Storage 16 GB microSD 32 GB microSD or NVMe SSD
Network 100 Mbps Ethernet or 2.4 GHz Wi-Fi Gigabit Ethernet or 5 GHz Wi-Fi

Operating System Requirements

Pyvorin Edge runs on Linux distributions with ARM64 kernels. We test primarily on Debian-derived systems because they have the best support for Raspberry Pi hardware, Python 3.12, and GCC toolchains.

Fully Supported Operating Systems

  • Raspberry Pi OS (64-bit) — Bookworm or later. This is the default operating system for Raspberry Pi and the one we recommend for beginners. It comes pre-configured with hardware acceleration, GPIO libraries, and a large community knowledge base.
  • Ubuntu Server 24.04 LTS (ARM64) — Excellent for headless deployments. Ubuntu's long-term support cycle makes it attractive for industrial installations. Ensure you install the python3.12-venv and build-essential packages.
  • Ubuntu Desktop 24.04 LTS (ARM64) — Suitable if you need a graphical interface on the gateway for local dashboard display.

Operating System Comparison

OS Python Version Kernel Best For
Raspberry Pi OS (64-bit) 3.11 (upgradeable to 3.12) 6.1+ Beginners, Pi-specific hardware
Ubuntu Server 24.04 3.12 (default) 6.8+ Enterprise, headless, LTS needs
Ubuntu Desktop 24.04 3.12 (default) 6.8+ Local GUI, kiosks, debugging
Debian 12 (Bookworm) 3.11 6.1+ Minimalist server setups

Python Requirements

Pyvorin Edge requires Python 3.12 or newer. This is non-negotiable because the codebase uses modern syntax features such as union types with the | operator, generic type parameter syntax, and improved tomllib support.

Checking Your Python Version


$ python3 --version
Python 3.12.3
  

If your system ships with Python 3.11 or earlier, you must install Python 3.12 before proceeding. On Raspberry Pi OS, you can build it from source or use the deadsnakes PPA on Ubuntu.

Installing Python 3.12 on Raspberry Pi OS


# Update package lists
sudo apt update

# Install dependencies for building Python
sudo apt install -y build-essential libncurses5-dev \
    libgdbm-dev libnss3-dev libssl-dev libreadline-dev \
    libffi-dev zlib1g-dev libsqlite3-dev wget

# Download and extract Python 3.12
wget https://www.python.org/ftp/python/3.12.3/Python-3.12.3.tgz
tar -xzf Python-3.12.3.tgz
cd Python-3.12.3

# Configure and build (this takes 20-40 minutes on a Pi 4)
./configure --enable-optimizations --prefix=/usr/local
make -j$(nproc)
sudo make altinstall

# Verify
python3.12 --version
  

Required Python Packages

Pyvorin Edge depends on the following packages, which are installed automatically via pip:

  • pyvorin — The core compiler and runtime (installed separately or from source).
  • paho-mqtt — Required only if you use MQTT ingest adapters.

No TensorFlow, PyTorch, or other heavy machine learning frameworks are required. This keeps the installation small and fast.

Memory and Disk Requirements

RAM Usage

Pyvorin Edge is memory-conscious. A typical deployment with 10 sensors, SQLite storage, and cloud sync uses between 80 MB and 150 MB of RAM. However, compilation tasks can spike memory usage because the C compiler needs working space.

Deployment Size Idle RAM Compilation Spike
Small (1-5 sensors) 60-90 MB Up to 300 MB
Medium (10-50 sensors) 100-150 MB Up to 500 MB
Large (50-200 sensors) 200-350 MB Up to 1 GB

For this reason, we recommend at least 4 GB of RAM if you plan to compile kernels on the gateway itself. If you only use pre-compiled .so files, 2 GB is sufficient even for medium deployments.

Disk Usage

Disk space is consumed by four things: the Python virtual environment, SQLite databases, compiled kernel libraries, and log files.

Component Typical Size Growth
Python venv + packages 150-300 MB Static
SQLite store (edge_store.db) 10-50 MB ~1 MB per 100k readings
Sync queue (sync_queue.db) 5-20 MB Depends on flush interval
Compiled kernels (.so files) 50-200 KB each Static after compilation
Logs 10-100 MB Rotated by configuration

A 16 GB microSD card is the absolute minimum. For production, use 32 GB or larger, and consider an external SSD if you expect high telemetry volumes or long retention periods.

Network Requirements

Pyvorin Edge does not require an always-on internet connection, but certain features need network access:

  • Local sensors (MQTT, HTTP). The gateway must be on the same network as your sensors or broker. No internet required.
  • Cloud sync. Outbound HTTPS (port 443) to your configured endpoint. No inbound ports need to be opened.
  • Health endpoint. If you query the health server remotely, port 8080 (or your configured port) must be reachable on the local network.
  • Installation. Downloading packages requires outbound access to PyPI and possibly GitHub.

Firewall Rules


# Outbound (required)
- TCP 443 to cloud endpoint (HTTPS cloud sync)
- TCP 443 to pypi.org, files.pythonhosted.org (installation)

# Inbound (optional, for local access)
- TCP 8080 on gateway LAN IP (health/metrics HTTP server)
- TCP 1883 on gateway LAN IP (if running local MQTT broker)
- TCP 8081+ on gateway LAN IP (if using HTTP adapter server)
  

Compiler Toolchain

To compile Python to native ARM64/NEON code, you need a working C compiler and Python development headers installed on the gateway.


# Raspberry Pi OS / Debian
sudo apt update
sudo apt install -y build-essential gcc g++ python3.12-dev libffi-dev

# Ubuntu 24.04
sudo apt update
sudo apt install -y build-essential gcc g++ python3.12-dev libffi-dev
  

If python3.12-dev is not available, use the version that matches your Python installation exactly:


python3 --version
# Then install the matching dev package
sudo apt install -y python3.12-dev
  

Summary Checklist

Before proceeding to installation, verify the following:

  • ARM64 device (Raspberry Pi 4/5 recommended)
  • 64-bit Linux OS (Raspberry Pi OS or Ubuntu 24.04)
  • Python 3.12 or newer installed and on PATH
  • At least 2 GB RAM (4 GB recommended)
  • At least 16 GB storage (32 GB recommended)
  • Network access for local sensors and optional cloud sync
  • build-essential, gcc, and python3.12-dev installed

If you can tick every box, you are ready to install Pyvorin Edge. Continue to the next article: Edge Installation.