AI in Dartford
- AI in Dartford: Server Configuration
This article details the server configuration powering the "AI in Dartford" initiative. It is aimed at new members of the system administration team and provides a comprehensive overview of the hardware, software, and network setup. Please read carefully and refer to related Internal Documentation for further details.
Overview
The "AI in Dartford" project utilizes a cluster of servers located in the Dartford data center. These servers are dedicated to machine learning tasks, specifically natural language processing (NLP) and computer vision. The primary goal is to analyze data related to Dartford Borough Council services and improve citizen engagement. The system employs a hybrid cloud approach, leveraging both on-premise hardware and cloud-based resources via Cloud Integration.
Hardware Configuration
The core of the system consists of five dedicated servers. Each server is built with high-performance components to handle the computationally intensive demands of AI workloads.
| Server Name | Role | CPU | RAM | Storage |
|---|---|---|---|---|
| dartford-ai-01 | Master Node (Kubernetes Control Plane) | Intel Xeon Gold 6248R (24 cores) | 256 GB DDR4 ECC | 2 x 1 TB NVMe SSD (RAID 1) |
| dartford-ai-02 | Worker Node (Model Training) | AMD EPYC 7763 (64 cores) | 512 GB DDR4 ECC | 4 x 2 TB NVMe SSD (RAID 10) |
| dartford-ai-03 | Worker Node (Model Training) | AMD EPYC 7763 (64 cores) | 512 GB DDR4 ECC | 4 x 2 TB NVMe SSD (RAID 10) |
| dartford-ai-04 | Inference Server | Intel Xeon Silver 4210 (10 cores) | 128 GB DDR4 ECC | 1 x 1 TB NVMe SSD |
| dartford-ai-05 | Data Storage & Backup | Dual Intel Xeon Silver 4208 (8 cores each) | 64 GB DDR4 ECC | 8 x 8 TB SAS HDD (RAID 6) |
All servers run on a dedicated 10Gbps network segment. Power redundancy is provided by dual power supplies and a UPS system, described in the Power Management document. Hardware monitoring is conducted via SNMP Monitoring.
Software Stack
The software stack is built around Kubernetes for container orchestration. This allows for efficient resource utilization and scalability.
| Component | Version | Description |
|---|---|---|
| Operating System | Ubuntu Server 22.04 LTS | Provides the base OS for all servers. |
| Kubernetes | v1.27.x | Container orchestration platform. See Kubernetes Documentation for details. |
| Docker | 20.10.x | Container runtime. |
| NVIDIA Drivers | 535.104.05 | Required for GPU acceleration. |
| TensorFlow | 2.12.x | Machine learning framework. |
| PyTorch | 2.0.x | Alternative machine learning framework. |
| Prometheus | 2.40.x | Monitoring system. |
All code is version controlled using Git Repository. The deployment pipeline is automated using CI/CD Pipeline. Security updates are managed via Automated Patching.
Network Configuration
The servers are connected to the internal network via a dedicated VLAN.
| Interface | IP Address | Subnet Mask | Gateway |
|---|---|---|---|
| eth0 (dartford-ai-01) | 192.168.10.10 | 255.255.255.0 | 192.168.10.1 |
| eth0 (dartford-ai-02) | 192.168.10.11 | 255.255.255.0 | 192.168.10.1 |
| eth0 (dartford-ai-03) | 192.168.10.12 | 255.255.255.0 | 192.168.10.1 |
| eth0 (dartford-ai-04) | 192.168.10.13 | 255.255.255.0 | 192.168.10.1 |
| eth0 (dartford-ai-05) | 192.168.10.14 | 255.255.255.0 | 192.168.10.1 |
Firewall rules are managed using Firewall Configuration. Access to the servers is restricted to authorized personnel only, as outlined in the Access Control Policy. Network performance is monitored via Network Monitoring Tools. The DNS configuration is detailed in DNS Records.
Future Considerations
Future plans include upgrading the GPUs on the worker nodes to NVIDIA A100s for increased performance. We are also exploring the integration of a dedicated model registry and versioning system using MLflow Integration. Further expansion of the storage capacity is anticipated based on data growth projections outlined in the Capacity Planning Report.
Server Room Access Emergency Procedures Data Backup Policy Security Audit Logs
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.* ⚠️