AI in Ashford
- AI in Ashford: Server Configuration
This document details the server configuration for the "AI in Ashford" project, a new initiative utilizing Artificial Intelligence for enhanced civic services. This guide is aimed at newcomers to the server administration team and provides a comprehensive overview of the hardware and software deployed. Please refer to the Server Administration Policy before making any changes.
Overview
The "AI in Ashford" project requires significant computational resources for model training, inference, and data storage. The system is built on a distributed architecture to ensure scalability and redundancy. We utilize a hybrid cloud approach, leveraging both on-premise hardware and cloud services from Cloud Provider X. This setup allows for flexibility and cost optimization. The core AI workloads are handled by GPU servers located in the Ashford Data Center, while less intensive tasks and long-term data storage are managed in the cloud. See the Network Topology Diagram for a visual representation of the system. Regular backups are performed according to the Backup and Recovery Procedures.
Hardware Specifications
The following tables detail the hardware specifications for the key server components. All servers are monitored using Monitoring System Y, and alerts are configured for critical metrics like CPU usage, memory consumption, and disk space. Physical security is managed by Security Team A.
GPU Servers (Ashford Data Center)
These servers are responsible for the bulk of the AI processing.
| Server Name | CPU | GPU | Memory (RAM) | Storage | Network Interface |
|---|---|---|---|---|---|
| ai-gpu-01 | Intel Xeon Gold 6248R @ 3.0GHz | NVIDIA A100 (80GB) x 2 | 512GB DDR4 ECC | 8TB NVMe SSD (RAID 0) | 100GbE |
| ai-gpu-02 | Intel Xeon Gold 6248R @ 3.0GHz | NVIDIA A100 (80GB) x 2 | 512GB DDR4 ECC | 8TB NVMe SSD (RAID 0) | 100GbE |
| ai-gpu-03 | AMD EPYC 7763 (64-Core) @ 2.45GHz | NVIDIA RTX 3090 (24GB) x 2 | 256GB DDR4 ECC | 4TB NVMe SSD (RAID 1) | 25GbE |
Database Servers (On-Premise)
These servers store the training data, model parameters, and processed results. They are subject to strict access control as detailed in the Access Control Policy.
| Server Name | CPU | Memory (RAM) | Storage | Database Software | Network Interface |
|---|---|---|---|---|---|
| db-primary | Intel Xeon Silver 4210 @ 2.1GHz | 128GB DDR4 ECC | 20TB SAS HDD (RAID 6) | PostgreSQL 13 | 10GbE |
| db-replica | Intel Xeon Silver 4210 @ 2.1GHz | 128GB DDR4 ECC | 20TB SAS HDD (RAID 6) | PostgreSQL 13 (Replica) | 10GbE |
Cloud Storage (Cloud Provider X)
This provides scalable storage for archiving and less frequently accessed data.
| Service | Storage Type | Capacity | Redundancy | Access Protocol |
|---|---|---|---|---|
| Cloud Archive | Glacier Deep Archive | 100TB | Geo-redundant | S3 |
| Cloud Backup | Standard S3 | 50TB | Geo-redundant | S3 |
Software Configuration
The software stack is based on Linux (Ubuntu Server 20.04 LTS) and utilizes containerization with Docker and orchestration with Kubernetes. All code is managed using Version Control System Z.
- **Operating System:** Ubuntu Server 20.04 LTS
- **Containerization:** Docker 20.10
- **Orchestration:** Kubernetes 1.23
- **Programming Languages:** Python 3.8, TensorFlow 2.7, PyTorch 1.10
- **Database:** PostgreSQL 13
- **Monitoring:** Monitoring System Y (Prometheus, Grafana)
- **Logging:** ELK Stack (Elasticsearch, Logstash, Kibana) – see Logging Configuration Guide.
- **Security:** Firewall Configuration, Intrusion Detection System.
Networking
The network is segmented into different zones for security purposes. The GPU servers are isolated on a dedicated VLAN. All traffic is monitored and filtered by the Firewall Configuration. Internal DNS is managed by DNS Server A. Refer to the Network Diagram for detailed network information.
Future Considerations
We plan to upgrade the GPU servers with newer generation GPUs (NVIDIA H100) in Q1 2024. We are also evaluating the use of New Technology B to further optimize performance. All changes will be documented in the Change Management System.
Server Documentation Index Troubleshooting Guide Security Best Practices Performance Tuning Guide
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.* ⚠️