AI in Suffolk
- AI in Suffolk: Server Configuration
This document details the server configuration supporting the "AI in Suffolk" project. It is intended for new system administrators and developers contributing to the project. This project focuses on utilizing Artificial Intelligence for agricultural optimization within the county of Suffolk.
Overview
The "AI in Suffolk" project requires significant computational resources for training machine learning models, processing sensor data from local farms, and serving prediction APIs. The infrastructure is primarily hosted on-premise for data security and low latency access to agricultural sensors. This document outlines the hardware, software, and networking configurations. We utilize a hybrid approach, with some data processing occurring on edge devices, and heavier lifting done on centralized servers. See also Data Pipeline Overview and Sensor Network Architecture.
Hardware Configuration
The core server infrastructure consists of three primary server types: Data Ingestion Servers, Model Training Servers, and API Servers. Each type is detailed below. Important consideration was given to power efficiency, as Suffolk County has a strong commitment to sustainable practices, as per the Suffolk Sustainability Plan.
Data Ingestion Servers
These servers are responsible for receiving data from various sensors deployed across farms in Suffolk County. They perform initial data validation and pre-processing before forwarding data to the model training servers.
| Component | Specification | Quantity |
|---|---|---|
| CPU | Intel Xeon Gold 6248R (24 cores) | 2 |
| RAM | 128GB DDR4 ECC Registered | 2 |
| Storage | 2 x 4TB NVMe SSD (RAID 1) | 2 |
| Network Interface | 10GbE | 2 |
| Power Supply | 800W Redundant | 2 |
Model Training Servers
These servers are equipped with powerful GPUs to accelerate the training of machine learning models. They leverage distributed training frameworks to handle large datasets. These servers are crucial for the Machine Learning Model Development process.
| Component | Specification | Quantity |
|---|---|---|
| CPU | AMD EPYC 7763 (64 cores) | 4 |
| RAM | 256GB DDR4 ECC Registered | 4 |
| GPU | NVIDIA A100 80GB | 8 |
| Storage | 4 x 8TB NVMe SSD (RAID 0) | 4 |
| Network Interface | 100GbE | 4 |
| Power Supply | 1600W Redundant | 4 |
API Servers
These servers host the APIs that provide access to the trained machine learning models. They handle requests from farmers and other applications. These servers are designed for high availability and scalability, as documented in API Scalability Plan.
| Component | Specification | Quantity |
|---|---|---|
| CPU | Intel Xeon Silver 4210 (10 cores) | 6 |
| RAM | 64GB DDR4 ECC Registered | 6 |
| Storage | 1 x 1TB NVMe SSD | 6 |
| Network Interface | 10GbE | 6 |
| Power Supply | 750W Redundant | 6 |
Software Configuration
The following software stack is used across the server infrastructure.
- Operating System: Ubuntu Server 22.04 LTS
- Containerization: Docker and Kubernetes for application deployment and orchestration. See Kubernetes Deployment Guide.
- Programming Languages: Python 3.9 is the primary language for data science and API development.
- Machine Learning Frameworks: TensorFlow and PyTorch are used for model training. Refer to Framework Comparison for details.
- Database: PostgreSQL is used for storing metadata, sensor data summaries, and model parameters. See Database Schema Documentation.
- Web Server: Nginx is used as a reverse proxy and load balancer for the API servers.
- Monitoring: Prometheus and Grafana are used for system monitoring and alerting. Refer to Monitoring Dashboard Setup.
Networking Configuration
The servers are connected via a dedicated 100GbE network. A firewall is in place to protect the infrastructure from unauthorized access. Network segmentation is implemented to isolate different parts of the system. Details of the network topology can be found in Network Diagram. We use VLANs to separate sensor data traffic from API traffic. The network is managed using Network Management System. A dedicated VPN connection is established for remote access.
Security Considerations
Security is paramount. All servers are hardened according to industry best practices. Regular security audits and vulnerability scans are conducted. Access control is strictly enforced using role-based access control (RBAC). Data encryption is used both in transit and at rest. See also the Security Policy Document.
Data Storage Solutions
Server Virtualization
Network Security Protocols
Database Administration Guide
Application Deployment Process
Kubernetes Configuration
Monitoring Tools
Firewall Configuration
IPv4 Addressing Scheme
VPN Setup Instructions
Backup and Disaster Recovery Plan
User Access Control
System Logging
Performance Tuning
Troubleshooting Guide
API Documentation
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.* ⚠️