AI in Bradford
- AI in Bradford: Server Configuration
This article details the server configuration supporting the "AI in Bradford" project, a local initiative leveraging artificial intelligence for urban planning and resource management. This guide is intended for new contributors to the wiki who may need to understand the underlying infrastructure.
Overview
The "AI in Bradford" project relies on a distributed server architecture to handle the large datasets and computational demands of machine learning models. The system is comprised of three key tiers: data ingestion, processing, and serving. Each tier utilizes specific hardware and software configurations, detailed below. The project utilizes Semantic MediaWiki extensions for data organization. Understanding Server Administration basics is crucial for maintaining this infrastructure. We adhere to strict Security Protocols throughout all tiers. Our network topology is documented in the Network Diagram section. Regular Backup Procedures are in place to prevent data loss.
Data Ingestion Tier
This tier is responsible for collecting data from various sources, including public APIs, sensor networks, and local databases. Data is validated, cleaned, and stored in a centralized data lake. We use Data Validation scripts to ensure data quality.
| Component | Specification | Quantity |
|---|---|---|
| Server Type | Dell PowerEdge R750 | 3 |
| Processor | Intel Xeon Gold 6338 | 3 per server |
| RAM | 256GB DDR4 ECC | 3 servers |
| Storage | 16TB RAID 6 SAS HDD | 3 servers |
| Network Interface | 10GbE | 3 |
The operating system of choice for this tier is Ubuntu Server 22.04 LTS. Data is initially staged in a Hadoop Distributed File System (HDFS) cluster before being moved to long-term storage. We utilize Kafka for real-time data streaming.
Processing Tier
The processing tier is the core of the AI system, where machine learning models are trained and evaluated. This tier requires significant computational power and utilizes GPU-accelerated servers. We employ Parallel Processing techniques to accelerate model training.
| Component | Specification | Quantity |
|---|---|---|
| Server Type | Supermicro SYS-220M-360 | 5 |
| Processor | AMD EPYC 7763 | 2 per server |
| RAM | 512GB DDR4 ECC | 5 servers |
| GPU | NVIDIA A100 (80GB) | 2 per server |
| Storage | 2TB NVMe SSD RAID 1 | 5 servers |
| Network Interface | 100GbE | 5 |
This tier runs Kubernetes for container orchestration, simplifying deployment and scaling of machine learning workloads. We use TensorFlow and PyTorch as our primary machine learning frameworks. Our Monitoring System continuously tracks GPU utilization. The Data Pipeline is crucial for efficient processing.
Serving Tier
The serving tier is responsible for deploying and serving trained machine learning models to end-users. This tier prioritizes low latency and high availability. We follow API Design best practices.
| Component | Specification | Quantity |
|---|---|---|
| Server Type | HP ProLiant DL380 Gen10 | 4 |
| Processor | Intel Xeon Silver 4310 | 2 per server |
| RAM | 128GB DDR4 ECC | 4 servers |
| Storage | 1TB NVMe SSD RAID 1 | 4 servers |
| Network Interface | 25GbE | 4 |
We utilize Docker containers to package and deploy the models. A Load Balancer distributes traffic across multiple server instances. The models are served via a REST API. We use Caching Mechanisms to improve response times. Database Integration is essential for model persistence. The system is integrated with the Bradford City Portal.
Network Diagram
A detailed network diagram outlining the connections between the three tiers and external data sources is available at Network Diagram. The diagram details the firewall rules and network segmentation.
Future Considerations
We are exploring the integration of Edge Computing to reduce latency and improve responsiveness. We are also investigating the use of Federated Learning to train models on distributed datasets without sharing sensitive data. We plan to implement a comprehensive Disaster Recovery Plan.
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.* ⚠️