AI in Cornwall
```wiki
- REDIRECT AI in Cornwall
AI in Cornwall: Server Configuration Overview
This article details the server configuration supporting the "AI in Cornwall" initiative. It aims to provide a comprehensive overview for new administrators and those seeking to understand the infrastructure. The project leverages a hybrid cloud approach, combining on-premise hardware with cloud-based services for scalability and resilience. This document will cover the core server components, network topology, and software stack. See also System Administration Guide for general MediaWiki site administration.
Core Infrastructure Components
The "AI in Cornwall" project relies on a tiered architecture. The core infrastructure consists of three primary tiers: Data Acquisition, Processing, and Delivery. Each tier has specific server requirements, outlined below. For information on Security Protocols, please see the dedicated security documentation.
| Tier | Purpose | Server Role | Key Technologies |
|---|---|---|---|
| Data Acquisition | Gathering data from sensors and external APIs | Edge Servers, Data Ingestion Servers | Python, MQTT, REST APIs, Database Schema |
| Processing | Performing AI/ML tasks, model training, and data analysis | GPU Servers, CPU Servers, Data Storage Servers | TensorFlow, PyTorch, CUDA, Data Analytics Tools |
| Delivery | Providing AI-driven insights and applications to end-users | Web Servers, Application Servers | Node.js, Flask, React, API Documentation |
On-Premise Hardware Specifications
The on-premise infrastructure is housed in a dedicated data center in Truro. The following tables detail the specifications of the key server types. Refer to Hardware Inventory for a complete listing of all equipment.
GPU Servers
These servers are critical for the computationally intensive tasks of model training and inference.
| Component | Specification |
|---|---|
| CPU | Dual Intel Xeon Gold 6248R (24 cores/48 threads per CPU) |
| GPU | 4 x NVIDIA A100 (80GB HBM2e) |
| RAM | 512GB DDR4 ECC Registered (3200 MHz) |
| Storage | 2 x 4TB NVMe SSD (RAID 0) + 2 x 16TB SAS HDD (RAID 1) |
| Network | Dual 100GbE NIC |
| Power Supply | 2 x 2000W Redundant Power Supplies |
Data Storage Servers
These servers provide high-capacity, reliable storage for datasets and model artifacts.
| Component | Specification |
|---|---|
| CPU | Intel Xeon Silver 4210 (10 cores/20 threads) |
| RAM | 256GB DDR4 ECC Registered (2666 MHz) |
| Storage | 8 x 20TB SAS HDD (RAID 6) - Total usable capacity: 120TB |
| Network | Quad 10GbE NIC |
| Power Supply | 2 x 850W Redundant Power Supplies |
Edge Servers
These servers are deployed closer to data sources (e.g., sensor networks) to perform initial data filtering and pre-processing.
| Component | Specification |
|---|---|
| CPU | Intel Core i7-8700K (6 cores/12 threads) |
| RAM | 32GB DDR4 (2666 MHz) |
| Storage | 1 x 1TB NVMe SSD |
| Network | Dual 1GbE NIC |
| Power Supply | 1 x 650W Power Supply |
Software Stack
The servers run a combination of Linux (Ubuntu Server 20.04 LTS) and containerized applications managed by Kubernetes. Key software components include:
- Operating System: Ubuntu Server 20.04 LTS
- Containerization: Docker, Kubernetes
- Programming Languages: Python 3.8, Node.js 16
- AI/ML Frameworks: TensorFlow 2.x, PyTorch 1.x
- Databases: PostgreSQL 13, Redis 6
- Message Queue: RabbitMQ 3.9
- Monitoring: Prometheus, Grafana. See Monitoring Dashboard Guide for details.
- Version Control: Git Repository Access
Network Topology
The on-premise network is a fully meshed topology with redundant connections. The servers are segmented into VLANs based on their roles. A firewall (Palo Alto Networks PA-Series) protects the network perimeter. The network is connected to the cloud provider (AWS) via a dedicated VPN connection. Refer to Network Diagrams for a visual representation of the network topology.
Cloud Integration
The "AI in Cornwall" project utilizes AWS for scalability and disaster recovery. Specifically:
- AWS S3: For storing large datasets and model backups.
- AWS EC2: For bursting capacity during peak demand.
- AWS SageMaker: For managed machine learning services.
- AWS RDS: For relational database instances. See Database Backup Procedures.
Future Considerations
Future plans include upgrading the GPU servers with newer NVIDIA H100 GPUs and exploring the use of serverless computing for certain tasks. We are also investigating the integration of federated learning to enable collaborative model training without sharing raw data. See the Roadmap for planned updates.
Data Center Management
Kubernetes Deployment
Security Best Practices
Database Administration
Network Troubleshooting
AWS Integration Guide
Monitoring and Alerting
Incident Response Plan
Backup and Recovery
System Performance Tuning
Software Updates
User Account Management
Firewall Configuration
Virtualization Technologies
Disaster Recovery Planning
Change Management Process
```
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.* ⚠️