Perf Distributed Testing

Overview

perf-distributed-testing provides production-ready Kubernetes Helm charts and Docker images for distributed load testing across JMeter, k6, and Gatling. Each tool uses its native distributed architecture — no compromise, no workarounds.

Deploy as many pods as your cluster can provide. Scale load horizontally with a single Helm value change.

Why Three Separate Charts?

JMeter, k6, and Gatling each have fundamentally incompatible distributed models:

Tool	K8s Workload	Coordination
JMeter	StatefulSet + Job	Java RMI (ports 1099, 50000)
k6	Deployment	None — stateless segment runners
Gatling	StatefulSet + Job	Akka remoting (port 2552)

A single unified chart cannot support all three without forcing every tool into an unnatural pattern. Each chart uses the right Kubernetes primitives for its tool.

Key Features

JMeter — RMI Controller-Workers

A StatefulSet gives workers predictable DNS names. The workerHosts Helm helper generates the exact -R flag from pod ordinals. An init container waits for all workers via DNS before the controller Job starts, so JMeter never attempts to connect to a pod that isn’t ready.

k6 — Stateless Segment Runners

k6 runners are stateless and interchangeable — a Deployment is the right fit. Each runner receives a RUNNER_INDEX and RUNNERS_TOTAL to calculate its independent slice of the virtual user load. Metrics stream to Prometheus or InfluxDB.

Gatling — Akka IP Discovery

Gatling uses IP addresses (not DNS names) for Akka remoting. An init container runs kubectl get pods to discover worker IPs and writes them to a shared volume. The master reads this file to form the Akka cluster. RBAC is created automatically.

Pricing Tiers

30-day money-back guarantee · Annual license · Apache-2.0 source

Quick Start

# 1. Get your free license key at martkos-it.co.uk/store/perf-distributed-testing-download
# 2. Get a registry token and create the imagePullSecret
TOKEN=$(curl -s https://updates.martkos-it.co.uk/api/v1/registry-token \
  -H "X-License-Key: YOUR_LICENSE_KEY" \
  -H "X-Product: perf-distributed-testing" | jq -r .token)

kubectl create secret docker-registry martkos-registry \
  --docker-server=registry.martkos-it.co.uk \
  --docker-username=your@email.com \
  --docker-password="$TOKEN" \
  --namespace perf-testing

# 3. Load your test plan
kubectl create configmap my-test-plan \
  --from-file=test-plan.jmx=./test.jmx \
  --namespace perf-testing

# 4. Deploy
helm install my-test oci://ghcr.io/markslilley/charts/perf-jmeter \
  --version 1.0.0 \
  --namespace perf-testing \
  --set test.configMapName=my-test-plan \
  --set worker.replicas=3 \
  --set global.imagePullSecrets[0].name=martkos-registry

FAQ

Do I need a license key to use the Helm charts?

The Helm charts are published to GHCR OCI and are publicly accessible — no license key required to pull them. The license key gates access to the pre-built Docker images from the private registry at registry.martkos-it.co.uk. You can also build images from source (Apache-2.0) without a key.

What’s the difference between Community and Enterprise?

Both tiers get identical technical access to all three charts and all three Docker images. Enterprise adds a priority support SLA (next business day response), architecture guidance, and a named contact at Martkos IT.

How many workers can I run?

Unlimited. There is no license-enforced cap on the number of pods, workers, or runners.

Which Kubernetes versions are supported?

Kubernetes 1.24+ is required for batch/v1 Job GA and stable StatefulSet DNS. Helm 3.10+ is required for helm test and OCI chart support.