AI in Bonaire
AI in Bonaire: Server Configuration and Deployment
This article details the server configuration for the "AI in Bonaire" project, a research initiative focused on applying artificial intelligence to marine conservation efforts around the island of Bonaire. This guide is intended for new system administrators and engineers joining the project, providing a comprehensive overview of the hardware and software stack.
Project Overview
The "AI in Bonaire" project utilizes machine learning algorithms to analyze underwater imagery and sensor data collected from autonomous underwater vehicles (AUVs) and stationary camera systems. This data is used to monitor coral reef health, identify invasive species, and track marine life populations. The system requires significant computational resources for data processing, model training, and real-time inference. This necessitates a robust and scalable server infrastructure. Refer to Data Acquisition for details on data sources and Machine Learning Models for a description of the algorithms employed.
Hardware Configuration
The server infrastructure is hosted in a dedicated, climate-controlled server room in Kralendijk, Bonaire. The primary components are outlined below. Power redundancy is provided by a UPS system and a backup generator, detailed in the Power and Cooling Infrastructure document. Network connectivity is provided by a dedicated fiber optic link, described in Network Architecture.
| Component | Specification | Quantity |
|---|---|---|
| CPU | Intel Xeon Gold 6248R (24 cores, 3.0 GHz) | 4 |
| RAM | 256 GB DDR4 ECC Registered 3200MHz | 4 |
| Storage (OS/Boot) | 500 GB NVMe SSD | 1 per server |
| Storage (Data) | 16 TB SAS 7.2k RPM HDD (RAID 6) | 2 arrays |
| GPU | NVIDIA Tesla V100 (32 GB HBM2) | 2 per server |
| Network Interface | 100 Gbps Ethernet | 2 per server |
Software Stack
The server environment is built on a Linux foundation, utilizing Ubuntu Server 22.04 LTS. Containerization is heavily employed for application isolation and scalability. The Software Deployment Guide provides detailed instructions on deploying applications.
Operating System
- Ubuntu Server 22.04 LTS (Kernel 5.15.0-76-generic)
- Systemd for system and service management
- Regular security updates managed via unattended upgrades
Containerization
- Docker 20.10.17
- Docker Compose v2.17.2
- Kubernetes 1.27 (for orchestration and scaling - see Kubernetes Configuration)
Data Management
- PostgreSQL 14.8 (for metadata and relational data)
- MinIO 2.0 (object storage for large datasets - see Data Storage Policies)
- NFS v4.1 (for shared storage between servers)
Machine Learning Frameworks
- Python 3.10
- TensorFlow 2.12
- PyTorch 2.0
- CUDA Toolkit 11.8 (for GPU acceleration)
- CuDNN 8.6
Server Roles & Configuration Details
The server infrastructure is divided into several roles, each with a specific purpose.
Data Ingestion Servers
These servers are responsible for receiving data from the AUVs and camera systems. They perform initial data validation and store the data in the MinIO object storage.
| Server Name | IP Address | Role | Key Software |
|---|---|---|---|
| data-ingest-01 | 192.168.1.10 | Data Ingestion | Docker, MinIO Client, Data Validation Scripts |
| data-ingest-02 | 192.168.1.11 | Data Ingestion (Backup) | Docker, MinIO Client, Data Validation Scripts |
Processing & Training Servers
These servers handle the computationally intensive tasks of data processing, model training, and hyperparameter tuning. They leverage the GPUs for accelerated computing. See Model Training Pipelines for more information.
| Server Name | IP Address | Role | Key Software |
|---|---|---|---|
| process-train-01 | 192.168.1.20 | Processing & Training | Docker, TensorFlow, PyTorch, CUDA, CuDNN |
| process-train-02 | 192.168.1.21 | Processing & Training | Docker, TensorFlow, PyTorch, CUDA, CuDNN |
Inference Servers
These servers deploy the trained machine learning models and provide real-time inference capabilities for analyzing incoming data streams. Real-time Inference Deployment details the deployment process.
| Server Name | IP Address | Role | Key Software |
|---|---|---|---|
| infer-01 | 192.168.1.30 | Inference | Docker, TensorFlow Serving, PyTorch Serve |
| infer-02 | 192.168.1.31 | Inference (Backup) | Docker, TensorFlow Serving, PyTorch Serve |
Security Considerations
Security is paramount. All servers are protected by a firewall and intrusion detection system (IDS). Access is restricted to authorized personnel only. See the Security Policy document for detailed information. Regular vulnerability scans are performed and software is kept up-to-date. Data encryption is used both in transit and at rest. Refer to Data Security Protocols for specific encryption methods.
Future Expansion
Planned future expansions include adding more GPU servers to handle increasing data volumes and more complex models. We are also investigating the use of distributed training frameworks to further accelerate model training. See Future Infrastructure Plans for details.
Main Page Data Acquisition Machine Learning Models Kubernetes Configuration Data Storage Policies Software Deployment Guide Model Training Pipelines Real-time Inference Deployment Power and Cooling Infrastructure Network Architecture Security Policy Data Security Protocols Troubleshooting Guide Monitoring and Alerting Backup and Recovery Contact Information
Intel-Based Server Configurations
| Configuration | Specifications | Benchmark |
|---|---|---|
| Core i7-6700K/7700 Server | 64 GB DDR4, NVMe SSD 2 x 512 GB | CPU Benchmark: 8046 |
| Core i7-8700 Server | 64 GB DDR4, NVMe SSD 2x1 TB | CPU Benchmark: 13124 |
| Core i9-9900K Server | 128 GB DDR4, NVMe SSD 2 x 1 TB | CPU Benchmark: 49969 |
| Core i9-13900 Server (64GB) | 64 GB RAM, 2x2 TB NVMe SSD | |
| Core i9-13900 Server (128GB) | 128 GB RAM, 2x2 TB NVMe SSD | |
| Core i5-13500 Server (64GB) | 64 GB RAM, 2x500 GB NVMe SSD | |
| Core i5-13500 Server (128GB) | 128 GB RAM, 2x500 GB NVMe SSD | |
| Core i5-13500 Workstation | 64 GB DDR5 RAM, 2 NVMe SSD, NVIDIA RTX 4000 |
AMD-Based Server Configurations
| Configuration | Specifications | Benchmark |
|---|---|---|
| Ryzen 5 3600 Server | 64 GB RAM, 2x480 GB NVMe | CPU Benchmark: 17849 |
| Ryzen 7 7700 Server | 64 GB DDR5 RAM, 2x1 TB NVMe | CPU Benchmark: 35224 |
| Ryzen 9 5950X Server | 128 GB RAM, 2x4 TB NVMe | CPU Benchmark: 46045 |
| Ryzen 9 7950X Server | 128 GB DDR5 ECC, 2x2 TB NVMe | CPU Benchmark: 63561 |
| EPYC 7502P Server (128GB/1TB) | 128 GB RAM, 1 TB NVMe | CPU Benchmark: 48021 |
| EPYC 7502P Server (128GB/2TB) | 128 GB RAM, 2 TB NVMe | CPU Benchmark: 48021 |
| EPYC 7502P Server (128GB/4TB) | 128 GB RAM, 2x2 TB NVMe | CPU Benchmark: 48021 |
| EPYC 7502P Server (256GB/1TB) | 256 GB RAM, 1 TB NVMe | CPU Benchmark: 48021 |
| EPYC 7502P Server (256GB/4TB) | 256 GB RAM, 2x2 TB NVMe | CPU Benchmark: 48021 |
| EPYC 9454P Server | 256 GB RAM, 2x2 TB NVMe |
Order Your Dedicated Server
Configure and order your ideal server configuration
Need Assistance?
- Telegram: @powervps Servers at a discounted price
⚠️ *Note: All benchmark scores are approximate and may vary based on configuration. Server availability subject to stock.* ⚠️