AI in Norwich
- AI in Norwich: Server Configuration
This document details the server configuration for the "AI in Norwich" project, a local initiative dedicated to exploring the applications of Artificial Intelligence within the city. This guide is aimed at new contributors to the wiki and provides a comprehensive overview of the hardware and software employed. Please read carefully before making any modifications.
Overview
The “AI in Norwich” project utilizes a cluster of servers located within the Norwich Research Park data centre. The servers are primarily used for machine learning model training, data analysis, and hosting web-based AI applications. The system is designed for scalability and redundancy, allowing for future growth and minimizing downtime. This configuration focuses on the core infrastructure. Additional information regarding Data Security Protocols and Network Topology can be found in separate articles.
Hardware Configuration
The cluster consists of three primary server nodes, designated as Node-A, Node-B, and Node-C. Each node is built with similar specifications to ensure consistency and simplify maintenance. A dedicated storage server handles data persistence.
| Server Node | CPU | RAM | Storage | Network Interface |
|---|---|---|---|---|
| Node-A | Intel Xeon Gold 6248R (24 cores) | 256 GB DDR4 ECC Registered | 2 x 4TB NVMe SSD (RAID 1) | 10 Gigabit Ethernet |
| Node-B | Intel Xeon Gold 6248R (24 cores) | 256 GB DDR4 ECC Registered | 2 x 4TB NVMe SSD (RAID 1) | 10 Gigabit Ethernet |
| Node-C | Intel Xeon Gold 6248R (24 cores) | 256 GB DDR4 ECC Registered | 2 x 4TB NVMe SSD (RAID 1) | 10 Gigabit Ethernet |
The storage server, designated 'Storage-1', provides centralized storage for all nodes.
| Component | Specification |
|---|---|
| Host Name | Storage-1 |
| CPU | Intel Xeon Silver 4210 (10 cores) |
| RAM | 64 GB DDR4 ECC Registered |
| Storage | 8 x 16TB SAS HDD (RAID 6) |
| Network Interface | 10 Gigabit Ethernet |
Power is supplied via redundant power supplies and a dedicated UPS system, detailed in the Power Management Documentation.
Software Configuration
All server nodes run Ubuntu Server 22.04 LTS. The software stack is designed to facilitate machine learning and data science workflows. The primary software components include Python 3.10, TensorFlow 2.12, PyTorch 2.0, and JupyterLab. A distributed file system, GlusterFS, is used to provide a unified namespace across the cluster.
| Software Component | Version | Purpose |
|---|---|---|
| Operating System | Ubuntu Server 22.04 LTS | Server Operating System |
| Python | 3.10.6 | Primary Programming Language |
| TensorFlow | 2.12.0 | Machine Learning Framework |
| PyTorch | 2.0.1 | Machine Learning Framework |
| JupyterLab | 3.5.0 | Interactive Development Environment |
| GlusterFS | 10.1 | Distributed File System |
Network Configuration
The servers are connected via a dedicated 10 Gigabit Ethernet network. Internal DNS resolution is managed by a local BIND server. Firewall rules are configured using `ufw` to restrict access to essential services. Refer to the Network Security Policy for detailed information. Each node is assigned a static IP address within the 192.168.1.0/24 subnet.
- Node-A: 192.168.1.10
- Node-B: 192.168.1.11
- Node-C: 192.168.1.12
- Storage-1: 192.168.1.20
The cluster utilizes a load balancer, configured with HAProxy, to distribute traffic to the active nodes.
Monitoring and Logging
Server health and performance are monitored using Prometheus and Grafana. Logs are collected and centralized using the ELK Stack (Elasticsearch, Logstash, Kibana). Alerts are configured to notify administrators of critical issues. Detailed logging configurations can be found in the Logging Standards Guide.
Future Considerations
Planned upgrades include migrating the storage server to NVMe SSDs for improved performance and exploring the use of containerization technologies like Docker and Kubernetes for application deployment. We are also investigating the addition of a dedicated GPU server for accelerated machine learning workloads.
Main Page Server Maintenance Procedures Troubleshooting Guide Data Backup Strategy User Access Control Security Audit Logs Software Update Schedule Hardware Inventory Contact Information Change Management Process Disaster Recovery Plan Incident Response Protocol Network Diagrams Virtualization Platform 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.* ⚠️