Server rental store

AI in Yorkshire

AI in Yorkshire: Server Configuration

This article details the server configuration used to support the “AI in Yorkshire” project, a research initiative focusing on applying artificial intelligence to regional data sets. This is intended as a guide for new system administrators and developers contributing to the project. Understanding the underlying infrastructure is crucial for effective deployment and maintenance. We will cover hardware specifications, software stack, networking, and security considerations. Please refer to the System Administration Guide for general MediaWiki server administration.

Hardware Overview

The core infrastructure consists of three primary server nodes: a head node for control and data processing, a storage node for large dataset management, and a GPU node dedicated to model training. All servers are located within a secure data center in Leeds. Detailed specifications are provided below.

Server Role Hostname CPU RAM Storage Network Interface
Head Node | ai-yorkshire-head.example.com | Intel Xeon Gold 6248R (24 cores) | 128GB DDR4 ECC | 1TB NVMe SSD (OS & Applications) + 4TB HDD (Temporary Data) | 10GbE |
Storage Node | ai-yorkshire-storage.example.com | AMD EPYC 7763 (64 cores) | 256GB DDR4 ECC | 120TB RAID6 HDD | 40GbE |
GPU Node | ai-yorkshire-gpu.example.com | Intel Xeon Gold 6338 (32 cores) | 256GB DDR4 ECC | 2TB NVMe SSD (OS & Models) | 10GbE |

These servers are powered by redundant power supplies and connected to a dedicated cooling system. See the Data Center Documentation for more information on the physical infrastructure. Regular hardware health checks are performed as outlined in the Server Maintenance Schedule.

Software Stack

The software environment is built around a Linux base, utilizing containerization for application deployment and management. We use Ubuntu Server 22.04 LTS as our operating system.

Component Version Purpose
Operating System | Ubuntu Server 22.04 LTS | Base Operating System | Docker | 23.0.1 | Containerization Platform | Kubernetes | 1.26 | Container Orchestration | Python | 3.10 | Primary Programming Language | TensorFlow | 2.12 | Machine Learning Framework | PyTorch | 2.0 | Machine Learning Framework | PostgreSQL | 15 | Database Management System | JupyterHub | 3.1 | Interactive Notebook Server |

All software is managed via a centralized configuration management system based on Ansible, ensuring consistency across all nodes. Software updates are applied according to the Release Management Policy. The Software Repository contains all custom software packages. We also utilize Virtual Environments for project specific dependencies.

Networking Configuration

The servers are connected via a dedicated VLAN within the data center network. Static IP addresses are assigned to each server, and DNS resolution is handled by an internal DNS server. Firewall rules are configured to restrict access to only necessary ports.

Server Role IP Address Subnet Mask Gateway DNS Server
Head Node | 192.168.1.10 | 255.255.255.0 | 192.168.1.1 | 192.168.1.1 | Storage Node | 192.168.1.20 | 255.255.255.0 | 192.168.1.1 | 192.168.1.1 | GPU Node | 192.168.1.30 | 255.255.255.0 | 192.168.1.1 | 192.168.1.1 |

Network monitoring is performed using Nagios, and alerts are configured for any network issues. The Network Diagram provides a visual representation of the network topology. Access to the servers is restricted via SSH Keys and multi-factor authentication. Please consult the Firewall Ruleset for detailed port access information.

Security Considerations

Security is paramount for the “AI in Yorkshire” project, given the sensitive nature of the data being processed. The following security measures are in place:

⚠️ *Note: All benchmark scores are approximate and may vary based on configuration. Server availability subject to stock.* ⚠️

View full article | Back to Server rental store