AI in South East England
AI in South East England - Server Configuration
This article details the server configuration supporting the "AI in South East England" project, a regional initiative showcasing Artificial Intelligence applications and research. It is aimed at newcomers to the wiki and provides a technical overview for those assisting with server maintenance or expansion. This configuration is designed for high availability, scalability, and security, catering to both data processing and web serving needs.
Overview
The "AI in South East England" project relies on a distributed server architecture hosted primarily within data centres in Oxfordshire and Hampshire. The system is broadly divided into three tiers: Web Front-End, Application/API Layer, and Data Storage/Processing. Each tier has specific hardware and software requirements, outlined below. The entire infrastructure is monitored using Nagios and Zabbix for proactive issue detection. Network infrastructure is managed using Cisco networking equipment.
Web Front-End Servers
These servers handle incoming web traffic, serving the project website and providing a user interface for accessing AI demonstrations and information. They are designed for high concurrency and responsiveness. We utilize Nginx as a reverse proxy and web server.
| Server Role | Hardware Specification | Operating System | Software Stack |
|---|---|---|---|
| Web Server 1 (oxf-web-01) | CPU: 2 x Intel Xeon Gold 6248R RAM: 128GB DDR4 ECC Storage: 2 x 1TB NVMe SSD (RAID 1) |
Ubuntu Server 22.04 LTS | Nginx 1.25.3, Let's Encrypt (HTTPS), PHP 8.2, MariaDB Client |
| Web Server 2 (oxf-web-02) | CPU: 2 x Intel Xeon Gold 6248R RAM: 128GB DDR4 ECC Storage: 2 x 1TB NVMe SSD (RAID 1) |
Ubuntu Server 22.04 LTS | Nginx 1.25.3, Let's Encrypt (HTTPS), PHP 8.2, MariaDB Client |
| Load Balancer (oxf-lb-01) | CPU: 1 x Intel Xeon E-2388G RAM: 32GB DDR4 ECC Storage: 2 x 500GB SSD (RAID 1) |
Ubuntu Server 22.04 LTS | HAProxy 2.6.4, Keepalived |
All web servers are configured with a Content Delivery Network (CDN) via Cloudflare to improve loading times for users across the South East and beyond. The website utilizes a custom MediaWiki skin.
Application & API Layer
This tier handles the core logic of the AI demonstrations and provides APIs for data access. Python is the primary programming language used, along with various AI frameworks. We utilize Docker and Kubernetes for containerization and orchestration.
| Server Role | Hardware Specification | Operating System | Software Stack |
|---|---|---|---|
| API Server 1 (ham-api-01) | CPU: 2 x AMD EPYC 7763 RAM: 256GB DDR4 ECC Storage: 4 x 2TB NVMe SSD (RAID 10) |
CentOS Stream 9 | Python 3.11, Flask, TensorFlow 2.12, PyTorch 2.0, Docker 24.0, Kubernetes 1.27 |
| API Server 2 (ham-api-02) | CPU: 2 x AMD EPYC 7763 RAM: 256GB DDR4 ECC Storage: 4 x 2TB NVMe SSD (RAID 10) |
CentOS Stream 9 | Python 3.11, Flask, TensorFlow 2.12, PyTorch 2.0, Docker 24.0, Kubernetes 1.27 |
| Message Queue (ham-mq-01) | CPU: 1 x Intel Xeon Silver 4310 RAM: 64GB DDR4 ECC Storage: 1 x 1TB SSD |
Ubuntu Server 22.04 LTS | RabbitMQ 3.9.x |
API security is managed via OAuth 2.0 and API key authentication. We leverage a microservices architecture using gRPC for inter-service communication.
Data Storage & Processing
This tier is responsible for storing and processing the large datasets used by the AI models. We utilize a combination of relational and NoSQL databases.
| Server Role | Hardware Specification | Operating System | Software Stack |
|---|---|---|---|
| Database Server 1 (oxf-db-01) | CPU: 2 x Intel Xeon Platinum 8380 RAM: 512GB DDR4 ECC Storage: 8 x 4TB SAS HDD (RAID 6) |
Red Hat Enterprise Linux 8 | PostgreSQL 15, pgAdmin |
| Database Server 2 (oxf-db-02) | CPU: 2 x Intel Xeon Platinum 8380 RAM: 512GB DDR4 ECC Storage: 8 x 4TB SAS HDD (RAID 6) |
Red Hat Enterprise Linux 8 | PostgreSQL 15, pgAdmin (Replication Server) |
| Data Processing Cluster (ham-dp-01 to ham-dp-05) | CPU: 1 x AMD Ryzen Threadripper PRO 5975WX (per node) RAM: 128GB DDR4 ECC (per node) Storage: 2 x 4TB NVMe SSD (per node) + 10TB NAS |
Ubuntu Server 22.04 LTS | Apache Spark 3.4.1, Hadoop 3.3.6, Jupyter Notebook, Python 3.11 |
Data backups are performed nightly to an offsite location using rsync and stored in a secure cloud storage solution. We employ Data encryption at rest and in transit. All servers are protected by a Firewall. The database utilizes PostGIS for geospatial data.
Network Configuration
The servers are interconnected via a dedicated 10Gbps network. Internal network addresses are allocated using a private IP range. External access is provided through a firewall with strict access control rules. DNS is managed using BIND.
Future Expansion
Planned expansions include the addition of GPU servers for accelerated machine learning tasks and integration with a cloud-based object storage service for long-term data archiving. We are also evaluating the use of Prometheus for advanced monitoring and alerting.
Special:Search/AI Special:Search/Server Special:Search/Configuration Special:Search/Database Special:Search/Networking Special:Search/Linux Special:Search/Python Special:Search/Kubernetes Special:Search/Docker Special:Search/Security Special:Search/Monitoring Special:Search/Backup Special:Search/PostgreSQL Special:Search/TensorFlow Special:Search/PyTorch
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.* ⚠️