Skip to content

Architecture Overview ​

Category: Architecture

The Azure Local Load Testing Framework is organised into five layers that take a declarative configuration file as input and produce validated, auditable test reports as output. Each layer has a clear responsibility boundary and communicates with adjacent layers through well-defined file contracts and PowerShell module APIs.


Five-Layer Stack ​

Solution Architecture


Layer Responsibilities ​

Layer 1 — Configuration ​

The configuration layer holds all cluster-specific, credentials, and workload parameters. A single variables.yml file acts as the source of truth.

ComponentLocationResponsibility
Master variables fileconfig/variables.ymlAll environment parameters, tagged by solution
Workload profilesconfig/profiles/<tool>/Per-tool YAML profile definitions
ConfigManager modulesrc/common/modules/ConfigManager/Filters, validates, and emits solution-scoped JSON
Schema validationconfig/schema/JSON Schema files that gate ConfigManager output

Layer 2 — Automation ​

The automation layer orchestrates pre-checks, installation, and execution across cluster nodes. All scripts consume only the ConfigManager-emitted JSON — never the raw YAML.

ComponentLocationResponsibility
Orchestrator scriptsscripts/*.ps1Top-level Start-* / Collect-* / Install-* entry points
Logger modulesrc/common/modules/Logger/Structured JSON-lines logging with correlation IDs
StateManager modulesrc/common/modules/StateManager/Checkpoint-based resume-after-failure
CredentialManager modulesrc/common/modules/CredentialManager/Key Vault, interactive, or parameter credential retrieval
Ansible rolessrc/ansible/roles/<tool>/Linux-target deployment (fio)

Layer 3 — Execution ​

The execution layer is the load-testing tools themselves, running on cluster nodes or inside guest VMs.

ToolTarget OSInstall Method
fioLinux VMsAnsible (Install-Fio.ps1)
iPerf3Linux nodesapt / dnf (manual)
HammerDBWindows nodesPowerShell remoting (Install-HammerDB.ps1)
stress-ngLinux nodesapt / dnf (manual)
VMFleetWindows (HCI host)Install-VMFleet.ps1

Layer 4 — Monitoring ​

The monitoring layer runs in parallel with execution, capturing Windows Performance Counter data and evaluating alert rules.

ComponentLocationResponsibility
MonitoringManager modulesrc/common/modules/MonitoringManager/PerfMon collection, Azure Monitor push
Alert rules filesmonitoring/<tool>/alert-rules.ymlPer-tool alert definitions
Grafana dashboardsmonitoring/dashboards/Real-time visualisation

Layer 5 — Reporting ​

The reporting layer aggregates raw JSON results, populates AsciiDoc templates, and renders final reports.

ComponentLocationResponsibility
ReportGenerator modulesrc/common/modules/ReportGenerator/Template population, asciidoctor-pdf invocation
Report templatesreports/templates/Per-tool AsciiDoc templates
Generated reportsreports/Output PDF / DOCX / XLSX files

Common Workflow Pattern ​

Every tool follows the same eight-phase lifecycle:

PhaseScript ActionStateManager Checkpoint
Pre-CheckValidate connectivity, prerequisitespre-check-complete
InstallDeploy tool binaries on target nodesinstall-complete
DeployConfigure test environmentdeploy-complete
TestExecute workload profiletest-complete
MonitorCollect PerfMon counters (parallel)monitor-complete
CollectParse and aggregate resultscollect-complete
ReportRender PDF/DOCX/XLSXreport-complete
Cleanup(Optional) Remove test artefactscleanup-complete

Any phase can be resumed from its checkpoint if the run is interrupted.


Security Model ​

  • Credentials are never hardcoded in scripts or configuration files
  • Three retrieval modes: Azure Key Vault (production), interactive prompt (development), parameter injection (CI/CD)
  • All credential access is logged with values masked
  • CI/CD pipelines use GitHub Secrets / Azure DevOps Service Connections

See Credential Management for implementation details.


Further Reading ​

Released under the MIT License.