Skip to content
Bitassert
Documentation

Performance Testing

Guides for authoring, running and analysing tests across web automation, API and performance — plus setup, collaboration and release notes.

Periodic Metrics

Periodic metrics are time-based snapshots of performance data collected during a single test run. They show how response time, errors, samples, and (for API) throughput change over the course of that execution — not across different runs. You view them on Performance → Test Results → run detail, in the chart sections below the summary cards and Aggregate Report.

What they are

Each interval stores a timestamped snapshot per API step or web action, plus a TOTAL row. Together they form a time series for that run — so you can see when response time, errors, or throughput shifted (for example during ramp-up or near the end), not just the final aggregate numbers.

Periodic metrics vs Aggregate Report

ConceptPeriodic metricsAggregate Report
ScopeOne run, over timeOne run, end-to-end totals
ViewLine charts over timeSummary table
PurposeSee when performance changedSee final overall numbers

Note:

Periodic metrics are not the same as — baseline marks a reference run across executions; periodic metrics trend within a single execution.

View periodic metrics — workflow

  1. Go to Performance > Test Results and open a run row (for example a completed API run).
  2. At the top, read run metadata and summary cards, then review the Aggregate Report for final per-step totals.
Performance result detail with summary cards and Aggregate Report table
  1. Scroll to Periodic Response Time Metrics. Use the metric dropdown (for example Average RT) and click legend lines to isolate a step. Review Response Codes Metrics beside the chart.
Periodic Response Time Metrics line chart and Response Codes donut chart
  1. Review Sample Error and Throughput Metrics — sample count, error %, and throughput per API step over time.
Sample Error and Throughput Metrics multi-series line chart over time
  1. Scroll further to CPU Usage Metrics and Memory Usage Metrics gauge charts for resource consumption during the run.
CPU Usage Metrics and Memory Usage Metrics gauge charts

Metrics captured per snapshot

API (per step label)

Samples (total/success/failure), average/min/max response time, P90/P95/P99 line RT, standard deviation, throughput (RPM), error rate %, received/sent KB per second, and response code counts.

Web (per action)

Action name, step name, samples, response times, percentiles, error rate %, and error type counts.

Live vs completed runs

Run stateBehavior
IN_PROGRESS / STOPPINGCharts update live as new periodic snapshots arrive
COMPLETED / ABORTEDFull periodic history loaded from the server; charts show the complete run timeline

Summary cards and the Aggregate Report use final aggregated values when the run is finished. Periodic charts always show the progression through the run.

Example

During a 10-minute load test, periodic charts may show response time rise at minute 2 (ramp-up), flatten at minutes 3–8, then spike at minute 9 when errors increase — patterns the single aggregate total alone would not reveal.

How this fits with other views

  • Periodic metrics → trends within one run over time
  • Aggregate Report → final per-step totals for that run
  • Baseline → which passed run is the reference standard for a release (see )
  • Test Results list → history of separate runs, not a built-in cross-run trend chart (see )

Expected outcome

You can open a performance run detail page and read how response time, errors, samples, and throughput evolved during that execution. Periodic charts answer what happened when inside a single run; the Aggregate Report answers what the final totals were.