Skii

System Overview

A hardware-integrated telemetry and analytics dashboard engineered for deployment on 32-bit ARM architectures (specifically Raspberry Pi OS arm/v7). The system interfaces directly with host Bluetooth hardware via BlueZ/DBus to aggregate sensor data, processes it utilizing an accelerated NumPy pipeline, and serves a real-time React-based user interface.

System Architecture

Backend (Data Acquisition & Processing)

• Runtime: Python 3.11
• Hardware Interfacing: Utilizes bleak for asynchronous Bluetooth Low Energy (BLE) communication, bridged to the host's bluetoothd via DBus socket mounting.
• Data Processing: Native array manipulation via numpy, accelerated by libopenblas0 to bypass standard QEMU/ARM emulation performance bottlenecks.
• OS Compliance: Compiled against libglib2.0-0t64 to ensure Y2038 compliance on 32-bit Linux distributions.

Frontend (User Interface)

• Framework: React
• Delivery: Statically compiled and served via a lightweight Nginx web server on port 80.

Infrastructure & Build Pipeline

The project utilizes a highly optimized, multi-stage Docker build pipeline designed to separate heavy compilation dependencies from the final production runtime.

• Target Architecture: linux/arm/v7
• Dependency Resolution: Leverages piwheels during the build stage to fetch pre-compiled ARM binaries, drastically reducing compilation overhead for heavy C-extensions.
• Artifact Generation: Images are packaged into a self-contained tarball (.tar.gz) to facilitate air-gapped or localized deployments without relying on a cloud-based container registry.

Deployment Protocol

Deployment requires moving the self-contained bundle to the target hardware, followed by an automated unpacking and execution script.

1. Transfer the Artifact: Transfer the skii-deployment.tar.gz archive from the build machine to the target Raspberry Pi utilizing any preferred file transfer method or physical storage medium.

2. Execute Deployment (On Target Hardware):

# Extract the deployment bundle
tar -xzvf skii-deployment.tar.gz

# Navigate to the extracted configuration directory
cd pi-deploy

# Initialize the stack (loads images into the local registry and provisions containers)
bash start.sh

Service Configuration (docker-compose.yml specs)

• Backend Privilege: Executes in network_mode: "host" with privileged: true to permit hardware-level BLE scanning.
• Volume Mapping: Requires - /var/run/dbus:/var/run/dbus to maintain the DBus bridge.
• Lifecycle: Both services operate under a restart: unless-stopped policy to ensure high availability following system reboots or fatal runtime exceptions.