Tokyo VPN Speed Monitor Dataset
Description
This dataset contains 1950+ automated measurements of VPN performance from Tokyo, Japan, collected period in December 2025. Key Features: - 15 VPN providers tested - Measurements every 6 hours (4 times per day) - Time-series data from early December 2025 - Statistical analysis with Coefficient of Variation (CV) for stability assessment Data Collection Period: - Start: Early December 2025 - Total measurements: 1950+ records Data Format: The CSV file contains the following columns: - Timestamp - VPN Service - Download (Mbps) - Upload (Mbps) - Ping (ms) - Instant Stability - Reliability (%) - Overall Score - Rank Methodology: Automated testing using Google Apps Script deployed on Google's free tier infrastructure. All measurements are performed from Tokyo, Japan, with testing occurring every 6 hours around the clock. Statistical analysis includes Coefficient of Variation (CV) calculation for stability assessment. System uptime: 99.8%. Operating cost: $0/month. Live System: - Dashboard: https://www.blstweb.jp/network/vpn/tokyo-vpn-speed-monitor/ - Source Code: https://github.com/hmy0210/vpn-stability-ranking - VPN Leak Detection Tool: https://www.blstweb.jp/network/vpn/security-diagnosis/ Use Cases: - VPN performance analysis and comparison - Network performance research - Time-series data analysis and forecasting - Educational purposes in networking courses - Benchmarking studies License: MIT License - Free to use for any purpose including commercial applications, research, and education.
Files
Steps to reproduce
Data Collection Setup: 1. Deploy Google Apps Script to Google's infrastructure 2. Configure VPN connections for 15 providers 3. Set up time-based triggers for 6-hour intervals 4. Configure Google Spreadsheet as data storage Measurement Process: 1. Script executes every 6 hours automatically 2. For each VPN provider: - Connect to VPN - Wait 30 seconds for connection stabilization - Measure download speed (3 trials) - Measure upload speed (3 trials) - Measure latency - Calculate average values - Disconnect from VPN 3. Record timestamp, VPN name, and measurement results 4. Calculate Coefficient of Variation (CV) for stability assessment 5. Store data in Google Spreadsheet Data Analysis: 1. Calculate CV = (Standard Deviation / Mean) × 100 2. Identify anomalies using 3-sigma method 3. Generate time-series visualizations 4. Export data as CSV for distribution Reproducibility: - All source code available on GitHub: https://github.com/hmy0210/vpn-stability-ranking - Measurement methodology documented - Testing environment: Tokyo, Japan (consistent location) - No manual intervention required after setup