AI in Bournemouth
AI in Bournemouth: Server Configuration
This article details the server configuration supporting the "AI in Bournemouth" project, a local initiative exploring applications of Artificial Intelligence within the Bournemouth, Poole, and Christchurch conurbation. This document is intended for new system administrators and developers contributing to the project. Understanding this configuration is crucial for maintaining stability, scalability, and security.
Overview
The "AI in Bournemouth" project leverages a hybrid server infrastructure, combining on-premise hardware with cloud-based resources. This allows for flexibility in handling varying workloads, from data preprocessing and model training to real-time inference. The on-premise servers are housed within a dedicated data centre at Bournemouth University. Cloud resources are primarily sourced from Amazon Web Services (AWS). A key component is the central data repository, using a Semantic MediaWiki extension for structured data storage.
On-Premise Server Infrastructure
The core on-premise infrastructure consists of three primary server roles: Database Server, Application Server, and File Server. These servers are interconnected via a dedicated Gigabit Ethernet network, with redundant connections for high availability.
Database Server
The Database Server is responsible for storing all project data, including datasets, model metadata, and user information. It’s critical for performance and reliability.
| Specification | Value |
|---|---|
| Server Model | Dell PowerEdge R740xd |
| Processor | 2 x Intel Xeon Gold 6248R (24 cores per processor) |
| RAM | 256GB DDR4 ECC Registered |
| Storage | 8 x 4TB SAS 12Gbps 7.2K RPM HDD (RAID 6) |
| Operating System | Ubuntu Server 22.04 LTS |
| Database System | MariaDB 10.6 |
The database is configured with replication for disaster recovery. Regular backups are performed using Bacula. Access to the database is strictly controlled via firewall rules and user authentication. The SQL queries are optimized for performance.
Application Server
The Application Server hosts the core AI applications and APIs. It handles user requests, processes data, and serves predictions.
| Specification | Value |
|---|---|
| Server Model | HP ProLiant DL380 Gen10 |
| Processor | 2 x Intel Xeon Silver 4210 (10 cores per processor) |
| RAM | 128GB DDR4 ECC Registered |
| Storage | 2 x 1TB NVMe SSD (RAID 1) |
| Operating System | CentOS Stream 9 |
| Web Server | Apache 2.4 |
| Programming Language | Python 3.9 |
The application utilizes a microservices architecture, deployed using Docker and orchestrated with Kubernetes. Load balancing is handled by a dedicated HAProxy instance.
File Server
The File Server provides central storage for large datasets and model files.
| Specification | Value |
|---|---|
| Server Model | Synology DiskStation DS1821+ |
| Processor | Intel Xeon J4125 |
| RAM | 16GB DDR4 |
| Storage | 8 x 16TB SATA HDD (RAID 5) |
| Operating System | Synology DSM 7.0 |
| File Protocol | NFS, SMB/CIFS |
Access to the File Server is managed through network shares and user permissions. Data is backed up regularly to an offsite location using rsync.
Cloud Infrastructure (AWS)
The project utilizes AWS for burst capacity and specialized services.
AWS Services
- Amazon EC2: Used for running machine learning training jobs, particularly those requiring GPUs. We utilize `p3.8xlarge` instances with NVIDIA V100 GPUs.
- Amazon S3: Provides scalable object storage for datasets and model artifacts.
- Amazon SageMaker: Used for model deployment and hosting, enabling real-time inference.
- Amazon RDS: Used as a secondary database for specific application components, providing a managed database service.
AWS EC2 Instance Specifications (Example)
| Specification | Value |
|---|---|
| Instance Type | p3.8xlarge |
| vCPUs | 64 |
| Memory (RAM) | 488 GB |
| Storage | 8 x 1 TB NVMe SSD |
| GPU | 4 x NVIDIA V100 |
| Operating System | Amazon Linux 2 |
Network Configuration
The network is segmented into three zones: Public, DMZ, and Private. The Application Server is located in the DMZ, providing a secure interface to the public internet. The Database Server and File Server are located in the Private zone, accessible only from the Application Server and authorized internal systems. Firewalls and Intrusion Detection Systems (IDS) are deployed to protect the network perimeter. VPN access is provided for remote administration.
Security Considerations
Security is a paramount concern. All servers are regularly patched and updated. Strong passwords and multi-factor authentication are enforced. Access control lists (ACLs) are used to restrict access to sensitive data. Regular security audits are conducted. We follow OWASP guidelines for web application security.
Future Enhancements
Planned enhancements include:
- Implementing a CI/CD pipeline for automated deployment.
- Migrating to a container orchestration platform like OpenShift.
- Exploring the use of machine learning for network intrusion detection.
- Integrating with a Monitoring System like Prometheus and Grafana.
Help:Contents Manual:Configuration Special:AllPages MediaWiki Server Administration
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.* ⚠️