You are here:
Trial Runs
Run small baseline tests to validate your JMeter or Playwright scripts and calculate target RPS before booking a full-scale load test.
Trial runs execute a small traffic simulation against your org to confirm that your scripts faithfully replicate production behavior and to size the actual test. A trial run with Available status within the last 15 days is required before you can book a scale test.
- In Scale Test, go to Test Environment Setup and select Manage Workloads. Enter your GitHub repository URL and personal access token, then click Fetch Workloads. Confirm imported workloads show Valid status.
- Select Trial Runs. Enter a trial name of your choice. Your org ID is auto-fetched from your logged-in org. No scripts are required. Click Submit.
- After the trial run completes, review accuracy rates across UI, DB Read, DB Write, and Error Logs. Rates of 90% or higher indicate a reliable production proxy. Trial runs show accuracy percentages only. Results files are generated for full scale tests only.
-
Calculate your target RPS.
- Production baseline available: Scale Test uses your trial run results to calculate accuracy and size the full test.
- No production baseline (for example, a new feature): Select a trial run date and time, then enter a scaling factor. Target RPS = base RPS × scaling factor ÷ 100%. To simulate 2x your current load, enter a scaling factor of 200%.
Best practices:
- Run it early. Don't wait until the week of your booked slot.
- Use production-representative data and the real persona profiles, permission sets, and licenses.
- If a production baseline is available and accuracy is below 90%, iterate on your workload scripts and data before retrying rather than rebooking. Trial runs for new features without a production baseline don't generate an accuracy metric.
- Confirm that the target RPS from your trial run matches the load you expect to achieve during your scale test.
Løste denne artikel dit problem?
Giv os besked, så vi kan forbedre os!
