AI in Lancashire
AI in Lancashire: Server Configuration
This article details the server configuration supporting Artificial Intelligence (AI) initiatives within Lancashire County Council. It is aimed at new system administrators and developers contributing to these projects. This document outlines the hardware, software, and network specifics required for optimal performance and scalability.
Overview
Lancashire County Council is increasingly utilizing AI for tasks ranging from predictive policing to resource allocation and citizen services. This requires a robust and scalable server infrastructure. The current setup employs a hybrid approach, utilizing both on-premise servers and cloud resources (primarily AWS) to balance cost, security, and performance. This document focuses on the on-premise components, with a brief overview of cloud integration. Refer to the Cloud Integration Guide for further details on AWS configuration.
Hardware Specifications
The primary AI processing is currently handled by a cluster of dedicated servers located within the Lancashire County Council data centre in Preston. These servers are designed for high-throughput computation and large dataset handling.
| Server Role | Model | CPU | RAM | Storage | Network Interface |
|---|---|---|---|---|---|
| AI Processing (Node 1-4) | Dell PowerEdge R750 | 2 x Intel Xeon Gold 6338 (32 cores/64 threads per CPU) | 512 GB DDR4 ECC REG | 4 x 4TB NVMe SSD (RAID 10) | 100 Gbps Ethernet |
| Data Storage (Node 5-6) | HPE ProLiant DL380 Gen10 | 2 x Intel Xeon Silver 4310 (12 cores/24 threads per CPU) | 256 GB DDR4 ECC REG | 8 x 16TB SAS HDD (RAID 6) | 25 Gbps Ethernet |
| Database Server | Lenovo ThinkSystem SR630 | 2 x AMD EPYC 7543P (32 cores/64 threads per CPU) | 256 GB DDR4 ECC REG | 2 x 1TB NVMe SSD (RAID 1) | 25 Gbps Ethernet |
These specifications are subject to change as AI workloads evolve. See the Hardware Refresh Policy for details. Network diagrams are available at Network Topology.
Software Stack
The software stack is carefully chosen to support the AI development and deployment pipeline. This includes the operating system, programming languages, AI frameworks, and database management system.
| Component | Version | Purpose |
|---|---|---|
| Operating System | Ubuntu Server 22.04 LTS | Server Operating System |
| Programming Languages | Python 3.10, R 4.3.1 | Primary languages for AI development |
| AI Frameworks | TensorFlow 2.12, PyTorch 2.0, scikit-learn 1.2.2 | Machine learning and deep learning frameworks |
| Database | PostgreSQL 15 | Data storage and management |
| Containerization | Docker 24.0.5, Kubernetes 1.27 | Application deployment and orchestration |
All software is kept up to date according to the Software Update Schedule. Security patching is prioritized and automated where possible. We utilize Ansible for configuration management.
Network Configuration
The AI server cluster resides on a dedicated VLAN within the Lancashire County Council network. This ensures isolation and security.
| Parameter | Value |
|---|---|
| VLAN ID | 200 |
| Subnet | 192.168.200.0/24 |
| Gateway | 192.168.200.1 |
| DNS Servers | 192.168.1.10, 192.168.1.11 |
| Firewall Rules | Allow inbound traffic on ports 22, 80, 443, 5432 from approved IP ranges. All other inbound traffic is blocked. |
Access to the servers is restricted to authorized personnel only, utilizing SSH key-based authentication. Detailed firewall rules are documented in the Firewall Configuration Document. Monitoring is handled via Nagios and Prometheus. Consider reviewing the Security Best Practices document before making any network changes.
Cloud Integration
As mentioned previously, a hybrid approach is employed. Cloud resources, hosted on AWS, are used for:
- Data backups and disaster recovery (using S3).
- Scaling compute resources for peak workloads (using EC2).
- Machine learning model training (using SageMaker).
Data transfer between on-premise servers and AWS is secured using VPN tunnels.
Future Considerations
Future plans include upgrading the GPU capabilities of the AI processing nodes to accelerate deep learning tasks. We are also exploring the use of more advanced container orchestration tools such as Nomad. Evaluating alternative database solutions like ClickHouse for handling time-series data is also being considered.
Main Page Server Documentation AI Project Overview Data Security Policy Incident Response Plan Change Management Process Disaster Recovery Plan Contact Information Frequently Asked Questions Troubleshooting Guide Performance Monitoring Capacity Planning Backup Procedures Software Licensing Hardware Inventory
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.* ⚠️