Skip to content

CI/CD · /// Release-Gated Performance Testing

Block performance regressions before they ship.

Wire LoadGen into the deploy pipeline. Trigger load tests from CI stages; gate deploys on HDX p95 drift, AVD login latency, API error rate, or per-step regression. Same scenario engine, every release.

Gate your pipeline on performanceSee CI/CD integration

Pipeline-triggered runsThreshold-gated deploysCitrix · AVD · Web · API

CI/CD pipeline — release-gated performance test flow.

The Problem

Performance regression catches up in production.

Most release pipelines gate on unit tests, integration tests, security scans. Performance testing typically runs separately — off-hours, on a different schedule — and regressions land in production before performance teams notice.

Off-hours testing misses release windows.

Performance tests scheduled weekly or monthly run against builds that have already shipped. Regression introduced in v1.2 surfaces after v1.3 has gone out.

Separate pipelines, separate decisions.

Functional tests gate the deploy; performance tests are an after-the-fact audit. The two systems drift; performance teams chase regressions instead of preventing them.

Threshold tuning is hard.

Hard pass / fail thresholds on absolute latency don’t survive normal variance. Smarter gating needs week-over-week comparison and AI-flagged anomalies — not static thresholds.

Why CI/CD-gated performance

Same engine. Triggered per release. Threshold-aware.

The same scenario engine that runs ad-hoc load tests triggers from pipeline stages. Per-step latency, p95 by protocol, error rate gated against the last good release. Analytics + AI flags regressions before the deploy unblocks.

Pipeline-triggered runs

CI stages trigger LoadGen runs via the published API. Jenkins, GitLab, GitHub Actions, Azure DevOps — any pipeline that can hit an HTTP endpoint can trigger a load test.

Threshold-gated deploys

Per-step latency, p95, error rate, assertion-failure thresholds gate the deploy. Block on regression vs the last good release — not on static absolute thresholds.

AI-flagged regression scoring

Analytics + AI scores each release against the baseline — surfaces what changed, where, and by how much. Threshold gating informed by anomaly detection, not just hard cutoffs.

Pipeline trigger

CI stage → LoadGen run → gate decision.

A CI stage posts to the LoadGen API; the scenario runs; the result returns to the pipeline. Deploy gated on threshold pass/fail — release blocked or unblocked based on measured data.

  • REST API trigger from any CI tool (Jenkins, GitLab, GitHub Actions, Azure DevOps).
  • Synchronous and asynchronous run modes — short scenarios block; long scenarios return run ID.
  • Threshold pass/fail returned to the pipeline as exit code.
  • Run history queryable via API for downstream reporting and dashboards.

CI/CD pipeline — stage triggers LoadGen run; gate on result.

Capability

What ships with CI/CD-gated performance testing.

API-triggered test runs

Pipeline stages POST to the LoadGen API to trigger a scenario. Run completes; status returned to the pipeline; deploy gated on the result.

Learn more

Per-step + p95 gating

Threshold gates on per-step latency, p95, error rate, assertion failures. Configurable per-scenario, per-environment, per-stage.

Learn more

Multi-test regression comparison

Each release compared against the last good run. Drill into Moments, Errors, per-step deltas — the regression source surfaces in the cockpit.

Learn more

Cross-platform coverage

Same engine drives Citrix HDX, AVD ARM, Horizon, RDS, Web, API regressions. One pipeline integration; all platforms gated.

Learn more

Alert on regression

When a release breaches the threshold, alerts route to the pipeline and to Email / Webhook / SMS / WhatsApp. Engineering knows immediately, not in the next standup.

Learn more

AI regression scoring

Analytics + AI scores each release vs baseline. Anomalies flagged automatically; engineering reviews the "what changed" report, not the full execution log.

Learn more

Regression comparison

This release vs the last good release.

Multi-test overlay compares the current release against the baseline. Per-step deltas, error-rate changes, p95 drift — the regression source surfaces in the cockpit.

  • Up to 5 runs overlaid on one chart — release history visual.
  • Drill into Moments, Errors, per-step measurements.
  • AI-flagged anomalies highlight what changed since the last good run.
  • Pass/fail decision informed by regression scoring, not just static thresholds.

Multi-test overlay — release-vs-baseline comparison.

Outcomes

What CI/CD-gated performance testing looks like.

Regression catch point

Before

Production

After

Pre-deploy gate

shift-left
Threshold logic

Before

Static

After

AI-informed

smarter
Test integration

Before

Separate audit

After

Pipeline-native

unified
Cross-platform gating

Before

Per platform

After

One engine

unified

Where CI/CD performance testing lives

Three pipeline-gating jobs.

EUC release gating

For Citrix / AVD-heavy estates, block deploys on HDX p95 drift, AVD login-time regression, or session-host density change vs the last good release.

See use case

API release gating

OpenAPI flows replay as load tests on every deploy. Block on per-endpoint latency regression, assertion failure, or error-rate threshold breach.

See use case

Multi-platform release gate

For organisations shipping web + EUC + API on the same release cycle, gate all three on the same scenario engine — one pipeline integration, full coverage.

See use case

Gate your pipeline on measured performance.

Bring your CI tool and a scenario; we’ll wire the trigger, configure the threshold gate, and walk through a regression-blocked deploy — on a call.

Gate your pipeline on performanceView pricing

Questions

Frequently asked.

Which CI tools does LoadGen integrate with?

LoadGen exposes a REST API for triggering and querying runs. Any CI tool that can POST to an HTTP endpoint — Jenkins, GitLab, GitHub Actions, Azure DevOps, CircleCI, TeamCity — can trigger and gate on LoadGen results.

How does threshold gating work?

Threshold rules configured per-scenario gate on per-step latency, p95, error rate, assertion failures. The pipeline gets a pass/fail exit code; downstream stages can branch on it.

Can the gate use AI-flagged regression scoring instead of static thresholds?

Yes. Analytics + AI scores each release against the baseline; the regression-score threshold can gate the deploy. Smarter than absolute pass/fail — surfaces anomalies that hard thresholds miss.

What if the scenario takes too long to run synchronously?

Async mode is supported. The CI stage triggers the run, gets a run ID, continues with downstream stages, and polls or webhook-receives the result. Long-running scenarios don’t block the pipeline.

Does this work for Citrix and AVD scenarios, or just web / API?

All of them. The same scenario engine drives Citrix HDX, AVD ARM, Horizon, RDS, Web, API. One pipeline integration; cross-platform gating.

What does CI/CD performance testing cost?

Load Testing tier is €1,099 per week at 50 vUsers; Monitoring is €899 per Agent per month. CI/CD-triggered runs are part of the Load Testing module — no separate license.

LoadGen Official Logo