Performance Testing Arrives: k6 and JMeter, Natively
TestRelic Team
Load testing has always lived somewhere else — a separate dashboard for k6, a different one for JMeter, neither talking to your functional test history. Today that changes: Performance Testing brings both into TestRelic as a first-class run type, sitting right next to your Playwright and Appium suites instead of off in a separate tool.
One document, one workspace
Whichever tool produced a run, TestRelic normalizes it at ingest into the same internal shape, so the workspace renders every run identically — you don't have to remember which format a given run came from, and neither does anyone you share it with.
tr upload k6 --summary summary.json --metrics metrics.json
tr upload jmeter --jtl results.jtlk6 uploads take the standard --summary-export JSON, with an optional --out json stream for finer-grained series. JMeter uploads take a JTL file in either CSV or XML.
A workspace built around finding the breach
Opening a run shows concurrency, throughput, p95 latency, and error rate as synchronized lanes on one timeline — with your latency gate drawn as a labelled line and any breach shaded directly on the lane. A failure-first playhead lands you on the primary breach automatically, and a moments list marks ramp-up, breach open/close, first failure, and peak load, so you're not scrubbing a five-minute run looking for the ten seconds that mattered.
Evidence is organized into tabs — Overview, Requests/Samples, Thresholds & Checks, Errors, and a Compare tab that stacks the run against the most recent passing run of the same scenario.
Service levels you set once
Define Gate (fail the run) and Warn (annotate it) thresholds per repository, scoped either to the whole run or to a one-minute sliding window — so a ten-second spike doesn't fail a run whose overall error rate is fine, and a slow overall creep doesn't hide behind one good minute. Thresholds are evaluated once, at ingest, so the verdict you see is stable no matter how you're viewing the run later.
Get started
Add the Performance section in your Repo Settings — tool cards for k6 and JMeter with ready-to-copy CI snippets, the service-level editor, baseline selection, and environment definitions. Then point your existing k6 or JMeter pipeline at the upload command above.
This is phase one. Streaming live k6 metrics, scheduled performance runs, per-endpoint SLOs, and additional tools are next.