AI in Serbia
- AI in Serbia: A Server Configuration Overview
This article details the server configuration supporting Artificial Intelligence (AI) initiatives within Serbia, focusing on the infrastructure deployed to facilitate research, development, and deployment of AI models. This guide is designed for newcomers to the MediaWiki platform and those interested in the technical underpinnings of AI infrastructure. Understanding the server configuration is critical for system administrators, data scientists, and machine learning engineers.
Overview
The AI infrastructure in Serbia is a distributed system comprised of several key components, designed for scalability and resilience. It leverages a combination of on-premise servers and cloud resources, primarily focusing on GPU-accelerated computing for training and inference. This setup aims to support various AI applications, including natural language processing, computer vision, and predictive analytics. The core philosophy is to provide researchers and developers with accessible and powerful resources to drive innovation in the field of AI. The current configuration supports both TensorFlow and PyTorch frameworks.
Hardware Specifications
The primary server cluster consists of the following hardware components. Detailed specifications are presented in the table below.
| Component | Specification | Quantity |
|---|---|---|
| CPU | Intel Xeon Gold 6248R (24 cores, 3.0 GHz) | 16 |
| RAM | 256 GB DDR4 ECC REG (3200 MHz) | 16 |
| GPU | NVIDIA A100 (80GB) | 8 |
| Storage (OS) | 1 TB NVMe SSD | 16 |
| Storage (Data) | 4 x 16 TB SAS HDD (RAID 6) | 4 Arrays |
| Network Interface | 100 GbE Mellanox ConnectX-6 | 16 |
These servers are housed in a dedicated data center facility with redundant power and cooling systems, ensuring high availability. The network infrastructure is designed to minimize latency and maximize bandwidth between servers. Network topology is a critical aspect of performance.
Software Stack
The software stack is built around a Linux operating system, specifically Ubuntu Server 22.04 LTS. This provides a stable and well-supported base for the AI applications.
| Software | Version | Purpose |
|---|---|---|
| Operating System | Ubuntu Server 22.04 LTS | Base OS |
| CUDA Toolkit | 12.2 | GPU Programming |
| cuDNN | 8.9.2 | Deep Neural Network Library |
| NVIDIA Driver | 535.104.05 | GPU Driver |
| Docker | 24.0.5 | Containerization |
| Kubernetes | 1.27.4 | Container Orchestration |
| TensorFlow | 2.13.0 | Machine Learning Framework |
| PyTorch | 2.0.1 | Machine Learning Framework |
Docker and Kubernetes are used for containerization and orchestration, allowing for easy deployment and scaling of AI models. The infrastructure also utilizes a version control system (Git) for managing code and configurations. Monitoring tools like Prometheus and Grafana are essential for tracking server performance.
Network Configuration
The network is configured using a flat network topology with a dedicated VLAN for AI traffic. This ensures isolation and security. The servers are connected via a high-speed 100 GbE network switch.
| Parameter | Value |
|---|---|
| Network Topology | Flat |
| VLAN ID | 100 |
| IP Addressing | Static |
| DNS Server | 8.8.8.8, 8.8.4.4 |
| Gateway | 192.168.1.1 |
Firewall rules are implemented to restrict access to the servers and protect against unauthorized access. Security protocols like SSH are used for remote administration. Regular network diagnostics are performed to ensure optimal performance. The entire network is monitored using intrusion detection systems.
Future Expansion
Plans are underway to expand the infrastructure with additional GPU servers and increased storage capacity. We are also exploring the use of specialized AI accelerators, such as TPUs, to further improve performance. Integration with cloud providers for burst capacity is also being considered. This will ensure that the infrastructure can continue to meet the growing demands of the AI community in Serbia. The expansion will also include better data backup solutions.
Server maintenance is scheduled regularly to ensure optimal operation.
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.* ⚠️