AI in the Sint Maarten Rainforest
AI in the Sint Maarten Rainforest: Server Configuration
This document details the server configuration supporting the "AI in the Sint Maarten Rainforest" project. This project utilizes artificial intelligence to analyze data collected from a network of sensors deployed within the rainforest, focusing on biodiversity monitoring and environmental change detection. This guide is intended for new contributors and system administrators maintaining this infrastructure. Understanding these configurations is crucial for effective System Administration and Data Analysis.
Overview
The system comprises three primary server roles: Data Acquisition, Processing & AI, and Data Storage. Each role is hosted on dedicated hardware to ensure optimal performance and scalability. The network topology is a star configuration, with all servers connecting to a central Network Switch for communication. Data security is paramount; all communication is encrypted using TLS/SSL. Regular Backups are performed and stored offsite. We utilize a Linux based operating system throughout the infrastructure.
Data Acquisition Servers
These servers are physically located close to the sensor network to minimize latency. They are responsible for receiving data from the sensor nodes, performing initial validation, and transmitting it to the Processing & AI servers. Each Data Acquisition server handles a specific geographic zone within the rainforest. We have two of these servers for redundancy.
| Specification | Value |
|---|---|
| Server Model | Dell PowerEdge R750 |
| CPU | Intel Xeon Silver 4310 (12 Cores) |
| RAM | 64GB DDR4 ECC |
| Storage | 1TB NVMe SSD (OS + Logging) |
| Network Interface | Dual 10GbE |
| Operating System | Ubuntu Server 22.04 LTS |
| Data Protocol | MQTT over TLS |
These servers run a custom-built Python script utilizing the Paho MQTT library to handle sensor data ingestion. Firewall rules are configured to allow only authorized connections from the sensor network and the Processing & AI servers. Monitoring is handled via Nagios.
Processing & AI Servers
These servers are the heart of the project, performing the complex computations required for data analysis and AI model execution. They receive validated data from the Data Acquisition servers, preprocess it, run AI models (specifically trained Convolutional Neural Networks for image analysis and Recurrent Neural Networks for time-series data), and store the results in the Data Storage servers. We employ a cluster of three Processing & AI servers for parallel processing.
| Specification | Value |
|---|---|
| Server Model | Supermicro SYS-2029U-TR4 |
| CPU | Dual Intel Xeon Gold 6338 (32 Cores per CPU) |
| RAM | 256GB DDR4 ECC |
| Storage | 2 x 2TB NVMe SSD (RAID 0 – for speed) |
| GPU | 2 x NVIDIA RTX A6000 (48GB VRAM each) |
| Network Interface | Dual 10GbE |
| Operating System | CentOS Stream 9 |
| AI Framework | TensorFlow 2.10 |
The AI models are deployed using Docker containers managed by Kubernetes. This allows for easy scaling and deployment of new model versions. Version Control is managed using Git. We utilize a specialized Monitoring Dashboard to observe GPU utilization and model performance.
Data Storage Servers
These servers are responsible for storing the raw sensor data, preprocessed data, and the results of the AI analysis. Data is stored in a PostgreSQL database with appropriate indexing for efficient querying. A secondary server is designated for offsite Disaster Recovery.
| Specification | Value |
|---|---|
| Server Model | HP ProLiant DL380 Gen10 |
| CPU | Intel Xeon Gold 5218 (16 Cores) |
| RAM | 128GB DDR4 ECC |
| Storage | 16 x 8TB SAS HDDs (RAID 6) |
| Network Interface | Dual 10GbE |
| Operating System | Debian 11 |
| Database | PostgreSQL 14 |
Data retention policies are defined to manage storage capacity. Regular Database Maintenance tasks are scheduled to ensure optimal performance. Access to the database is strictly controlled using Access Control Lists. The servers leverage Data Compression techniques to reduce storage requirements.
Networking
All servers are connected via a dedicated VLAN. A Load Balancer distributes traffic across the Processing & AI servers. The network is monitored using Zabbix for performance and security alerts.
Future Considerations
We are exploring the integration of Edge Computing to reduce latency and bandwidth requirements. We are also evaluating the use of Serverless Computing for certain AI tasks. Further Security Audits are planned to ensure the ongoing protection of our data.
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.* ⚠️