AI in Brighton and Hove
- AI in Brighton and Hove: Server Configuration Overview
This article details the server configuration supporting the "AI in Brighton and Hove" initiative. It's aimed at newcomers to this wiki and provides a technical overview of the hardware and software used. This system is designed to handle large datasets related to local AI projects and provide a platform for model training and deployment. We will cover hardware specifications, software stack, and networking details. It is vital to understand the system architecture before contributing to or modifying the infrastructure. Please refer to the System Administration Guidelines for general wiki policies.
Hardware Infrastructure
The core infrastructure consists of three primary server nodes, each serving a specific role. These are located in a secure data center facility within Brighton and Hove. Detailed specifications are provided below. Consider reviewing the Data Center Security Protocols before physical access is requested.
| Server Node | Role | CPU | RAM | Storage |
|---|---|---|---|---|
| Node 1 (Athena) | Primary Data Storage & Pre-processing | Dual Intel Xeon Gold 6338 | 256GB DDR4 ECC | 6 x 18TB SAS HDDs (RAID 6) + 2 x 1TB NVMe SSDs (OS/Cache) |
| Node 2 (Hermes) | Model Training & Development | AMD EPYC 7763 | 512GB DDR4 ECC | 4 x 8TB SAS HDDs (RAID 10) + 2 x 2TB NVMe SSDs (OS/Cache) + 4 x NVIDIA RTX A6000 GPUs |
| Node 3 (Zeus) | Model Deployment & API Server | Intel Xeon Silver 4310 | 128GB DDR4 ECC | 2 x 4TB SAS HDDs (RAID 1) + 1 x 1TB NVMe SSD (OS/Cache) |
All servers are connected via a dedicated 10GbE network. Power redundancy is provided by dual power supplies and an uninterruptible power supply (UPS). Refer to the Power Management Documentation for further details. We also utilise a dedicated Backup and Disaster Recovery system.
Software Stack
The software stack is built around a Linux operating system, providing a stable and flexible platform. We utilize Debian 11 (Bullseye) as the base OS. The following components are essential to the system's operation.
| Component | Version | Purpose |
|---|---|---|
| Operating System | Debian 11 (Bullseye) | Base OS |
| Python | 3.9 | Primary programming language for AI/ML tasks |
| TensorFlow | 2.8.0 | Deep learning framework |
| PyTorch | 1.10.0 | Deep learning framework |
| Docker | 20.10.7 | Containerization platform |
| Kubernetes | 1.23.4 | Container orchestration |
| PostgreSQL | 14 | Database for metadata and results |
All code is version controlled using Git and hosted on a private GitLab instance. Software updates are managed through a controlled release process detailed in the Software Release Management document. Security patches are applied immediately as per the Security Incident Response Plan.
Networking Configuration
The network is segmented to enhance security and performance. Each server node has a static IP address within the 192.168.1.0/24 subnet. A dedicated VLAN is used for communication between the server nodes and the external API gateway.
| Server Node | IP Address | VLAN | Firewall Rules |
|---|---|---|---|
| Node 1 (Athena) | 192.168.1.10 | 10 | Allows inbound connections from Node 2 & Node 3 on specific ports (PostgreSQL, SSH) |
| Node 2 (Hermes) | 192.168.1.11 | 10 | Allows inbound connections from Node 1 & Node 3 on specific ports (TensorBoard, SSH) |
| Node 3 (Zeus) | 192.168.1.12 | 20 | Allows inbound connections from external API gateway on port 80/443 |
Firewall rules are managed using `iptables` and are regularly reviewed and updated. See the Network Security Policy for more information. Internal DNS resolution is provided by a local bind9 server. We also utilize Intrusion Detection Systems for security monitoring. Further details on network topology can be found in the Network Diagram.
Future Considerations
We are evaluating the integration of a distributed storage system, such as Ceph, to improve scalability and resilience. We are also exploring the use of more powerful GPUs to accelerate model training. The Roadmap for AI Infrastructure details these planned upgrades. Consider the System Monitoring Procedures for operational stability.
Main Page
Server Administration
Data Storage Policies
Security Best Practices
Performance Tuning Guide
Troubleshooting Guide
Git Workflow
Database Administration
API Documentation
Monitoring Tools
Patch Management
Incident Reporting
Capacity Planning
User Access Control
System Logs
Contact Information
Change Management Process
Virtualization Strategy
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.* ⚠️