AI in Cumbria
AI in Cumbria: Server Configuration
This document details the server configuration for the "AI in Cumbria" project, a distributed computing initiative focused on analyzing environmental data collected within the county of Cumbria, England. This guide is intended for new system administrators and developers contributing to the project. It outlines the hardware, software, and network configurations necessary for optimal performance and reliability. Understanding these details is crucial for maintaining and expanding the system. We will cover hardware specifications, software stack, network topology, and security considerations. Please refer to the Main Page for project overview and associated documentation.
Hardware Overview
The "AI in Cumbria" project utilizes a hybrid server infrastructure, combining on-premise servers at the University of Cumbria with cloud-based resources from a dedicated AWS Virtual Private Cloud (VPC). This design allows for both high-performance processing and scalability. The on-premise servers handle initial data ingestion and pre-processing, while the cloud resources handle the computationally intensive machine learning tasks. See Data Flow Diagram for a visual representation.
The core on-premise server specifications are detailed below:
| Component | Specification | Quantity |
|---|---|---|
| CPU | Intel Xeon Gold 6248R (24 cores/48 threads) | 3 |
| RAM | 256 GB DDR4 ECC Registered | 3 |
| Storage (OS) | 500GB NVMe SSD | 3 |
| Storage (Data) | 16TB SAS HDD (RAID 6) | 1 (Shared) |
| Network Interface | Dual 10GbE | 3 |
| Power Supply | Redundant 1200W Platinum | 3 |
The AWS infrastructure consists of EC2 instances configured as follows:
| Instance Type | Quantity | Configuration |
|---|---|---|
| p3.8xlarge | 10 | NVIDIA V100 GPUs, 32 vCPUs, 244 GB RAM |
| r5.large | 5 | 2 vCPUs, 8 GB RAM (for control plane & monitoring) |
| s3 | N/A | 100TB Data Storage (Object Storage) |
Refer to Hardware Maintenance Procedures for detailed information on server maintenance.
Software Stack
The software stack is designed to facilitate data ingestion, processing, and model deployment. We utilize a Linux-based operating system (Ubuntu Server 22.04 LTS) across all servers. Containerization is achieved using Docker and orchestration via Kubernetes. See Software Installation Guide for detailed installation instructions.
Here’s a breakdown of the key software components:
| Software | Version | Purpose |
|---|---|---|
| Ubuntu Server | 22.04 LTS | Operating System |
| Python | 3.9 | Primary Programming Language |
| TensorFlow | 2.12 | Machine Learning Framework |
| PyTorch | 2.0 | Machine Learning Framework |
| Docker | 24.0 | Containerization Platform |
| Kubernetes | 1.27 | Container Orchestration |
| PostgreSQL | 15 | Database for Metadata and Results |
We also employ Prometheus for monitoring and Grafana for visualization. The API Documentation provides details on interacting with the AI models.
Network Configuration
The network topology consists of a dedicated VLAN for the "AI in Cumbria" project, both on-premise and within the AWS VPC. On-premise servers connect to the university network via 10GbE switches. The AWS VPC is connected to the university network via a secure VPN tunnel using OpenVPN. This allows for secure data transfer between the on-premise and cloud resources.
- **On-Premise VLAN:** 192.168.10.0/24
- **AWS VPC CIDR:** 10.0.0.0/16
- **VPN Tunnel:** IPSec with AES-256 encryption. Refer to the Network Diagram for a detailed illustration.
Security Considerations
Security is paramount. The following measures are in place:
- **Firewall:** A strict firewall policy is enforced on both the on-premise servers and within the AWS VPC, allowing only necessary traffic.
- **Access Control:** Role-Based Access Control (RBAC) is implemented using Kubernetes to restrict access to resources. See Access Control List.
- **Data Encryption:** All data in transit is encrypted using TLS/SSL. Data at rest is encrypted using AES-256 encryption within the PostgreSQL database and S3 buckets.
- **Regular Security Audits:** Regular security audits are conducted to identify and address vulnerabilities. See Security Audit Logs.
- **Intrusion Detection System (IDS):** An IDS is deployed to monitor network traffic for malicious activity. Refer to IDS Configuration.
Future Expansion
Future plans include integrating additional data sources and expanding the cloud infrastructure to accommodate increasing computational demands. The Roadmap document details these plans. We are also exploring the use of Federated Learning to enable distributed model training without sharing sensitive data. Please contribute to the Issue Tracker with suggestions and bug reports. Finally, see Contact Information for assistance.
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.* ⚠️