AI in the Curaçao Rainforest
AI in the Curaçao Rainforest: Server Configuration
This article details the server configuration deployed to support the "AI in the Curaçao Rainforest" project, a research initiative focused on biodiversity monitoring using artificial intelligence. This guide is intended for newcomers to our MediaWiki site and provides a technical overview of the hardware and software infrastructure. Understanding this setup is crucial for anyone contributing to the project's data analysis or system maintenance.
Project Overview
The "AI in the Curaçao Rainforest" project utilizes a network of camera traps and audio recorders deployed throughout the Shete Boka National Park in Curaçao. Data collected from these devices is streamed to our central server for processing, analysis, and long-term storage. The AI models employed are primarily focused on species identification from images and soundscapes. This requires significant computational resources and a robust, reliable server infrastructure. Our data pipeline is designed for scalability and resilience. We also use a version control system to manage all configurations.
Server Hardware
The core of our infrastructure consists of three primary servers: a primary data ingestion server, a processing server, and a storage server. Each server is located in a secure, climate-controlled data center in Willemstad. Detailed specifications are provided below:
| Server Role | CPU | RAM | Storage | Network Interface |
|---|
| Data Ingestion Server | Intel Xeon Gold 6248R (24 cores) | 128 GB DDR4 ECC | 2 x 4TB NVMe SSD (RAID 1) | 10 Gbps Ethernet | |
| Processing Server | 2 x AMD EPYC 7763 (64 cores total) | 256 GB DDR4 ECC | 4 x 8TB SAS HDD (RAID 5) + 2 x 1TB NVMe SSD (OS) | 25 Gbps Ethernet | |
| Storage Server | Supermicro Ultra Storage Server | 128 GB DDR4 ECC | 32 x 16TB SAS HDD (RAID 6) | 40 Gbps InfiniBand | |
These servers are connected via a dedicated internal network. Power redundancy is provided by dual power supplies and an uninterruptible power supply (UPS). Regular system backups are performed to an offsite location. We monitor server health using Nagios.
Software Configuration
The software stack is built around a Linux foundation, utilizing Ubuntu Server 22.04 LTS as the operating system for all servers. Key software components include:
- Data Ingestion Server: Nginx web server, Rsync for data transfer, and a custom Python script for initial data validation. This server also hosts the API endpoint for camera trap uploads.
- Processing Server: Python 3.10, TensorFlow 2.12, PyTorch 2.0, CUDA Toolkit 11.8 (for GPU acceleration), and Jupyter Notebooks for development and experimentation. We utilize a containerization strategy with Docker to ensure reproducibility.
- Storage Server: Ceph distributed storage system for scalability and data redundancy. This provides a unified namespace for all project data. Data archiving is handled through Ceph's lifecycle management policies.
- Regular security audits.
- Strong password policies.
- Two-factor authentication for all administrative accounts.
- Intrusion detection system (IDS).
- Data encryption at rest and in transit.
- Regular security patching of all software.
- Expanding the storage capacity of the Storage Server.
- Adding a dedicated GPU server for faster model training.
- Implementing a more sophisticated monitoring system.
- Automating the deployment of new AI models.
- Integrating with a cloud storage provider for disaster recovery.
- Telegram: @powervps Servers at a discounted price
Network Topology
The network infrastructure is designed for high bandwidth and low latency. The following table outlines key network components:
| Component | IP Address | Function |
|---|
| Router | 192.168.1.1 | Gateway to the internet, firewall | |
| Data Ingestion Server | 192.168.1.10 | Receives data from camera traps | |
| Processing Server | 192.168.1.20 | Runs AI models | |
| Storage Server | 192.168.1.30 | Stores all project data | |
| Monitoring Server | 192.168.1.40 | Runs Nagios and other monitoring tools | |
All communication between servers is secured using SSH and TLS. We employ a firewall configuration to restrict access to only necessary ports. Network performance is regularly monitored using Iperf3.
AI Model Deployment
AI models are deployed on the Processing Server using Docker containers. Each model is encapsulated in its own container, ensuring isolation and reproducibility. The models are served via a REST API using Flask. We use model versioning to track changes and ensure traceability. The AI models are regularly retrained using the latest data, following a continuous integration/continuous deployment (CI/CD) pipeline.
Security Considerations
Security is paramount. The following measures are in place:
Future Enhancements
Planned future enhancements include:
Server Resource Allocation
The following table details the resource allocation for key processes:
| Process | Server | CPU Allocation | Memory Allocation |
|---|---|---|---|
| Data Ingestion | Data Ingestion Server | 4 cores | 16 GB | | Model Training | Processing Server | 32 cores | 128 GB | | Model Serving | Processing Server | 8 cores | 64 GB | | Ceph Storage | Storage Server | 16 cores | 64 GB | |
Contact the administrators for any questions or concerns regarding the server infrastructure.
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 configurationNeed Assistance?
⚠️ *Note: All benchmark scores are approximate and may vary based on configuration. Server availability subject to stock.* ⚠️