4 Best Ways to Run Global Latency Tests
From one-click browser checks to CI pipeline integration, there's a method for every workflow.
Method 1: One-Click Web Check
The most straightforward way is to open the speed test homepage, enter the target address, and select the desired node group. It's ideal for quick troubleshooting and demonstrating issues to non-technical colleagues. The result page link can be shared directly, and the recipient can view the full data upon opening.
Method 2: Batch Testing
When you need to compare dozens or hundreds of targets at once, batch Ping, batch Tcping, and batch HTTP(S) allow you to submit up to 256 targets in a single run, with support for IP ranges and CIDR format expansion. The results table can be sorted by any node and filtered by response rate, and you can copy the data as a table with one click.
Method 3: REST API Integration
Integrate latency testing into your release process: after each deployment, call the API to run global checks on the new version's endpoints, and automatically roll back if latency or packet loss exceeds thresholds. The API returns structured JSON, and integrating with any CI system requires only a dozen lines of script.
Method 4: Continuous Monitoring
For critical domains, we recommend setting up website monitoring to probe from global nodes continuously at intervals from one minute to one hour. Data is retained for 90 days, allowing you to review latency curves at any point in time. With the newly launched email notifications, you'll know about failures within the first minute.