MaxoPerf platform glossary
This page explains the terms MaxoPerf uses in its console, API, and documentation. With them you can find your way around the product, configure tests correctly, and read the results.
A test in MaxoPerf is a saved configuration that describes what to run: the engine (Taurus, JMeter, k6), the uploaded script or YAML, the load profile, the target location(s), and any failure criteria. A test is the template; a run is a single execution of that template.
In MaxoPerf: you manage tests in the Tests list of a project. One test can have many runs over time, so you can compare results across builds or dates.
A run is a single execution of a test at a point in time. It has a start time, an end time, a status (running, completed, failed, aborted), and a full set of result metrics captured during execution.
In MaxoPerf: each run gets a unique URL. Share it with your team to discuss specific results. The run-detail page shows the load profile timeline, per-label latency and throughput charts, error breakdown, and any failure criteria evaluations.
Runner
Section titled “Runner”A runner is a compute instance (a container or VM) that runs the test engine and generates load against the target. MaxoPerf manages the runner lifecycle: it provisions runners before the run, monitors them during it, and deprovisions them after.
In MaxoPerf: for managed-cloud tests, MaxoPerf allocates runners automatically from the selected location. For BYOC tests, runners run in your own infrastructure through the Outpost agent.
Location
Section titled “Location”A location is a geographic or network region where MaxoPerf runners execute your test. Test from the location closest to your target (or from several locations at once) to get results closer to real user experience.
In MaxoPerf: the location selector is in the test configuration. MaxoPerf operates the managed locations. Private locations use your own Outpost. A test can target multiple locations in one run for distributed load generation.
Outpost
Section titled “Outpost”An Outpost is a MaxoPerf agent that you deploy inside your own network (on-premises, private cloud, or VPC) to create a private location. Traffic from the Outpost never leaves your infrastructure before it reaches the target. You need this to test services that are not publicly reachable.
In MaxoPerf: install the Outpost agent with the Helm chart or the provided Docker Compose bundle. After it registers, the Outpost appears as a selectable location in the test configuration. The Outpost polls MaxoPerf for run assignments and executes them locally.
See also: BYOC for cloud-hosted private runners.
BYOC (Bring Your Own Cloud)
Section titled “BYOC (Bring Your Own Cloud)”BYOC lets you run MaxoPerf test runners in your own cloud account (AWS, GCP, Azure, or similar) instead of on MaxoPerf-managed infrastructure. Your data, traffic, and credentials stay in your account. MaxoPerf only orchestrates the run.
See the canonical card: BYOC in the SEO Glossary.
In MaxoPerf: configure BYOC by connecting your cloud account in workspace settings. After that, BYOC locations appear next to managed locations in the test configuration. BYOC is the recommended approach for regulated industries, or when target systems are only reachable from inside your cloud perimeter.
Workspace
Section titled “Workspace”A workspace is the top-level organizational unit in MaxoPerf. It maps to a company, team, or billing entity. All projects, tests, runs, and integrations belong to a workspace.
In MaxoPerf: you invite users to workspaces with role-based permissions. Billing is per workspace. API keys and integrations are scoped to a workspace.
Project
Section titled “Project”A project groups related tests within a workspace. Projects typically map to a service, application, or team. You can have many projects in one workspace.
In MaxoPerf: tests, runs, and result dashboards live in a project. You can set access control per project within a workspace.
Failure criteria
Section titled “Failure criteria”Failure criteria are pass/fail rules that MaxoPerf evaluates at the end of (or during) a run. If any criterion is breached, MaxoPerf marks the run as failed. This lets you automate performance testing in CI: the build fails if the performance gate is not met.
Common failure criterion types:
- Latency percentile exceeds threshold (e.g.
p95 > 1000ms) - Error rate exceeds threshold (e.g.
error rate > 1%) - Throughput falls below minimum (e.g.
RPS < 100) - Apdex score below target (e.g.
Apdex < 0.8)
In MaxoPerf: add failure criteria in the test configuration. They appear on the run-detail page under the “Criteria” tab, with each criterion’s measured value and pass/fail status.
Run artifacts
Section titled “Run artifacts”Run artifacts are the files MaxoPerf stores from a run: raw results, logs, JMeter reports, engine output, and any custom exports. You can download them after the run completes.
See the canonical card: Run artifacts in the SEO Glossary.
In MaxoPerf: the run-detail “Artifacts” tab lists them. Use them for detailed post-run analysis, to share raw data with external tools, or to archive results.
Distributed load test
Section titled “Distributed load test”A distributed load test runs load generators on multiple runner instances or in multiple locations at once, and combines the results into a single run view. This is how MaxoPerf scales tests beyond what a single runner can generate.
In MaxoPerf: select multiple runners or multiple locations in the test configuration. The run detail page shows both the aggregate view and per-location breakdowns.
Test script
Section titled “Test script”A test script is the file (or files) you upload to define the test behavior: a
Taurus YAML, JMeter .jmx, k6 JavaScript, or similar engine-specific format. The script
defines scenarios, request sequences, parametrization, and assertions.
In MaxoPerf: you upload test scripts per test, and MaxoPerf versions them. You can update the script and re-run to compare the new behavior with previous runs.
Where to go next
Section titled “Where to go next”- Engines and tooling glossary: Taurus, JMeter, k6 concepts.
- Cookbook: failure criteria pass/fail gates: how to set up CI-gated tests.
- BYOC in the SEO Glossary: canonical BYOC card.
- Run artifacts in the SEO Glossary: canonical artifacts card.