Explore test results

Once the parsing job completes, test results are available for analysis. The solution provides comprehensive metrics and tools to help you understand your application’s performance under load.

Test run summary metrics

When a test completes, the solution generates a summary that includes the following metrics:

• Average response time - The average response time, in seconds, for all requests generated by the test.

• Average latency - The average latency, in seconds, for all requests generated by the test.

• Average connection time - The average time, in seconds, it takes to connect to the host for all requests.

• Average bandwidth - The average bandwidth for all requests generated by the test.

• Total count - The total number of requests.

• Success count - The total number of successful requests.

• Error count - The total number of errors.

• Requests per second - The average requests per second for all requests generated by the test.

• Percentiles - Response time percentiles including p50 (median), p90, p95, and p99, showing the distribution of response times across all requests.

Test runs table

The test runs table displays all historical test runs for a scenario. You can:

• View summary metrics for each test run.

• Set a baseline test run for performance comparison.

• Download the table as a CSV file.

• Toggle columns to customize your view.

• Select a test run to view detailed results.

Baseline comparison

You can designate a test run as a baseline to compare future test runs against it. When a baseline is set:

• The test runs table shows percentage differences (+/-%) compared to the baseline for each metric.

• The baseline indicator helps you quickly identify performance improvements or regressions.

• You can change or clear the baseline at any time.

Detailed test results

Selecting a test run opens the detailed results view with three tabs: Test Run Results, Errors, and Artifacts.

Baseline information

If a baseline test run is set, it displays at the top of the page. You can choose Show Actual, Show Percentage, or Remove Baseline to control how baseline comparisons are displayed.

Test Run Results table

The results table provides detailed metrics with the following features:

Dimension views

Toggle between three views using the dimension buttons:

• Overall - Aggregated results across all endpoints and regions

• By Endpoint - Results broken down by individual endpoints

• By Region - Results broken down by AWS region

Action buttons

• Show Actual - Display actual metric values

• Show Percentage - Display percentage differences from baseline

• Remove Baseline - Clear the baseline comparison

Data export and customization

• Download the results table as a CSV file

• Toggle columns to customize your view

• Filter and sort data to focus on specific metrics

• Filter and sort data to focus on specific metrics.

Errors tab

The errors tab provides detailed error analysis:

• View error counts by type.

• See errors aggregated by overall test or by endpoint.

• Identify patterns in failed requests.

• Troubleshoot issues with specific endpoints or regions.

Artifacts tab

The artifacts tab allows you to access all files generated during the test run:

• View individual artifacts (logs, results files).

• Download specific artifacts for offline analysis.

• Download all test run artifacts as a single archive.

S3 results structure

In version 4.0, the S3 results structure has changed to improve organization:

• New structure - scenario-id/test-run-id/results-files.

• Legacy structure - Tests run before version 4.0 show all result files at the scenario ID level.

Note

Test results are displayed in the console. You can also access the raw test results directly in the Amazon S3 bucket under the Results folder. For more information on Taurus test results, see Generating Test Reports in the Taurus User Manual.