AI in the Indonesian Rainforest
AI in the Indonesian Rainforest: Server Configuration
This article details the server configuration utilized for the "AI in the Indonesian Rainforest" project. This project leverages machine learning to analyze audio data collected from remote sensors deployed within the Indonesian rainforest, focusing on biodiversity monitoring and illegal logging detection. This guide is intended for new contributors and system administrators managing this infrastructure.
Project Overview
The "AI in the Indonesian Rainforest" project requires robust, reliable, and scalable server infrastructure. The core components involve data ingestion from various sensor nodes, pre-processing, model training, inference, and data visualization. The servers are geographically distributed – a primary cluster located in Jakarta for core processing and a smaller, redundant cluster in Singapore for disaster recovery. The system is designed to handle large volumes of audio data, demanding significant computational resources and storage capacity. We utilize a hybrid cloud approach, leveraging both on-premise hardware and cloud services. See Data Acquisition Strategy for details on the sensor network.
Hardware Configuration (Jakarta Primary Cluster)
The Jakarta cluster hosts the majority of the processing workload. It consists of the following servers:
| Server Role | Server Name | CPU | RAM | Storage | Network Interface |
|---|---|---|---|---|---|
| Data Ingestion & Pre-processing | data-ingest-01 | 2 x Intel Xeon Gold 6248R (24 cores/48 threads) | 256 GB DDR4 ECC REG | 6 x 4TB SAS 12Gbps 7.2K RPM | 10GbE |
| Model Training (GPU) | train-gpu-01 | 2 x Intel Xeon Gold 6338 (32 cores/64 threads) | 512 GB DDR4 ECC REG | 2 x 2TB NVMe PCIe Gen4 SSD (OS & Temp) + 16 x 8TB SAS 12Gbps 7.2K RPM (Data) | 100GbE |
| Model Training (GPU) | train-gpu-02 | 2 x Intel Xeon Gold 6338 (32 cores/64 threads) | 512 GB DDR4 ECC REG | 2 x 2TB NVMe PCIe Gen4 SSD (OS & Temp) + 16 x 8TB SAS 12Gbps 7.2K RPM (Data) | 100GbE |
| Inference & API | inference-01 | 2 x Intel Xeon Silver 4310 (12 cores/24 threads) | 128 GB DDR4 ECC REG | 2 x 1TB NVMe PCIe Gen3 SSD | 10GbE |
| Database Server (PostgreSQL) | db-master | 2 x Intel Xeon Gold 6230 (20 cores/40 threads) | 256 GB DDR4 ECC REG | 8 x 4TB SAS 12Gbps 7.2K RPM (RAID 10) | 10GbE |
These servers are interconnected via a dedicated, low-latency network. Detailed specifications are available in Hardware Inventory. Power redundancy is handled by dual UPS systems, and cooling is provided by a dedicated CRAC unit.
Software Stack
The Jakarta cluster utilizes the following software stack:
- Operating System: Ubuntu Server 22.04 LTS
- Containerization: Docker and Kubernetes
- Programming Languages: Python 3.9, with libraries such as TensorFlow, PyTorch, and Librosa.
- Database: PostgreSQL 14
- Message Queue: RabbitMQ
- API Framework: Flask
- Monitoring: Prometheus and Grafana – see Monitoring and Alerting.
Hardware Configuration (Singapore Disaster Recovery Cluster)
The Singapore cluster provides redundancy in case of a major outage in Jakarta. It is a scaled-down version of the primary cluster.
| Server Role | Server Name | CPU | RAM | Storage | Network Interface |
|---|---|---|---|---|---|
| Data Ingestion & Pre-processing | dr-ingest-01 | 1 x Intel Xeon Silver 4310 (12 cores/24 threads) | 128 GB DDR4 ECC REG | 4 x 4TB SAS 12Gbps 7.2K RPM | 10GbE |
| Inference & API | dr-inference-01 | 1 x Intel Xeon Silver 4310 (12 cores/24 threads) | 64 GB DDR4 ECC REG | 1 x 1TB NVMe PCIe Gen3 SSD | 10GbE |
| Database Server (PostgreSQL - Replica) | dr-db-replica | 1 x Intel Xeon Silver 4310 (12 cores/24 threads) | 128 GB DDR4 ECC REG | 4 x 4TB SAS 12Gbps 7.2K RPM (RAID 10) | 10GbE |
Data replication between the Jakarta and Singapore databases is handled using PostgreSQL streaming replication. Failover procedures are documented in the Disaster Recovery Plan. The Singapore cluster is kept in a warm-standby state, ready to take over critical services within a defined Recovery Time Objective (RTO).
Network Topology
The network topology is a hybrid setup. Internal servers communicate via a dedicated VLAN. External access to the API is secured via a reverse proxy and firewall. Connectivity between Jakarta and Singapore is established via a dedicated VPN tunnel. See Network Diagram for a visual representation.
| Component | IP Address Range | Subnet Mask |
|---|---|---|
| Jakarta Internal Network | 192.168.10.0/24 | 255.255.255.0 |
| Singapore Internal Network | 192.168.20.0/24 | 255.255.255.0 |
| VPN Tunnel | 10.0.0.0/30 | 255.255.255.252 |
Future Considerations
Future enhancements include:
- Expanding the GPU cluster for faster model training.
- Implementing automated scaling based on workload demand.
- Integrating a more sophisticated data lake solution for long-term storage and analysis. Refer to Data Storage Strategy for details.
- Exploring the use of edge computing to reduce latency and bandwidth requirements. See Edge Computing Implementation.
Security Considerations are paramount. Regular security audits and vulnerability scans are conducted to ensure the integrity and confidentiality of the data.
Main Page Project Documentation Data Pipeline Model Training Procedures API Documentation Troubleshooting Guide Contact Us Server Maintenance Schedule Software Updates Data Privacy Policy Compliance Regulations Backup and Restore Procedures System Logs Incident Response 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.* ⚠️