Skip to content

Repository files navigation

🚀 AI Workstation & Performance Monitor v7.0

License: MIT C++ Standard Platform

AI Workstation & Performance Monitor is a professional-grade telemetry and diagnostic suite engineered for high-performance Windows systems. Designed for local AI development (LLMs, Stable Diffusion), heavy engineering workloads, and enthusiast gaming, this tool provides deep hardware visibility and performance intelligence.


📑 Table of Contents


🏗️ Key Features

📊 Performance Analytics Hub

A dedicated suite for real-time tracking of system performance during intensive workloads.

  • Session Telemetry: Captures FPS distribution, CPU/GPU pressure, and thermal signatures at 1Hz.
  • Latency Monitoring: Integrated network latency (Ping) tracking to diagnose connection stability.
  • Visual Timelines: Clear line charts showing simultaneous hardware utilization trends.
  • Proactive Alerts: Color-coded status cards that flag thermal stress or system bottlenecks.

🧠 System Intelligence Core

An advanced heuristic analysis engine that provides deep diagnostic insights.

  • Bottleneck Identification: Real-time detection of CPU, GPU, or VRAM limitations.
  • Intelligent Recommendations: Actionable insights for system tuning and optimization.
  • Thermal Heuristics: Analyzes temperature patterns to detect potential hardware throttling.

⚡ Energy & Economics

  • Real-time Wattage: Detailed component-level power draw breakdown.
  • Operational Cost: Session-based electricity cost tracking with monthly projections.

🧠 AI & ML Workstation Dashboard

Optimized for the modern AI developer, supporting over 30+ identification profiles.

  • AI Stack Detection: Automatically recognizes PyTorch, TensorFlow, Ollama, and Stable Diffusion environments.
  • VRAM Capacity Estimator: Calculates compatibility for 22+ popular LLMs (Llama 3, SDXL, etc.) based on hardware memory.
  • Env Validation: Verification of CUDA readiness and Python availability.

📊 Data Interoperability

v7.0 introduces professional exports for data-driven performance tuning:

  • JSON Export: Deep session metrics formatted for automated analysis pipelines.
  • CSV Export: High-resolution timeline data (Timestamp, FPS, Latency, CPU, GPU).
  • Professional Reports: Structured session summaries for quick performance debriefing.

🛠️ Technical Architecture

Hardware Abstraction Layer (HAL)

The modular HAL Architecture (IGpuProvider) decouples the core UI from specific vendor APIs, ensuring a stable and extensible framework.

  • Native Support: NVIDIA NVML, Windows PDH, DXGI.
  • Optimized Engine: Asynchronous polling ensures monitoring has zero impact on primary workloads.

🚀 Installation & Build

Prerequisites

  • Windows 10/11
  • CMake 3.15+
  • Visual Studio 2019/2022

Quick Start

cmake -B build
cmake --build build --config Release
./build/Release/SystemMonitor.exe

📄 License

This project is licensed under the MIT License.


Optimized for High-Performance Windows Workstations.

About

A high-performance C++ System Monitor featuring GPU compute metrics, AI workload detection, and real-time energy cost tracking.

Topics

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages