liquidware-ninjaSecond brain & toolkit
Personal project. Research and learning only. Not for production. No warranty.Not for production
  • Built on my own time for research and learning. Not affiliated with, endorsed by or supported by my employer, Liquidware or Microsoft.
  • Bring your own Liquidware licenses. Don't run any of it in production or against tenants and subscriptions that matter. Test in a dedicated lab first.
  • Provided "as is", without warranty of any kind, under each repository's license. You use it at your own risk and own the Azure charges it incurs.
  • Contributions and forks are welcome and encouraged on public repos. See Contribute.

ux-assess

High-level overview of the private Liquidware Stratusphere UX assessment and deliverable pipeline: pulls telemetry from the Hub API (or Inspector exports), verifies collection health, normalizes and HMAC-pseudonymizes five canonical tables, scores platform-aware SpotCheck thresholds, classifies applications for Base Image vs. FlexApp delivery, and generates multi-format deliverables from one set of findings.

Private Companion RepoPython 3.10+PowerShell 7Stratusphere /lwl/apiPower BI PBIP (TMDL/PBIR)

Five-Stage Pipeline Architecture

StageWhat It DoesKey Design Choice
1. Intake & Collection HealthConfigures engagement metadata (current-state, performance, sizing, or migration-baseline) and compares Hub machines against expected inventory (active, dropped_off, sparse, inventory_only, missing_from_hub).Catches silent Connector ID Key callback failures before sizing or performance conclusions are drawn.
2. Pull or ImportQueries https://<hub>/lwl/api?json=... across Inspectors 0, 1, 2, 3, 4, and 10 with explicit business-hour slices, or ingests offline Inspector .xlsx/.csv exports when the Hub API is unreachable.Defaults to a dry-run Plan mode that prints every query before making network calls; enforces TLS certificate trust.
3. Normalize & PseudonymizeMaps Hub columns into five canonical tables (machines, machine_hourly, app_usage, logins, ux_daily) and hashes user/machine identifiers with keyed HMAC-SHA256.Same customer-held HMAC key across before and after runs preserves cohort joins without exporting raw PII.
4. Analyze (SpotCheck & Sizing)Evaluates business-hour p95 metrics against platform-aware SpotCheck thresholds, computes right-sized compute tiers (Light, Medium, Heavy, Power), and tiers applications for Base Image vs. FlexApp layering.Missing or unmapped columns are explicitly reported as "not measured" rather than silently passing checks.
5. Multi-Format DeliverablesRenders an Excel workbook (.xlsx), PDF report (.pdf), PowerPoint deck (.pptx), and a text-based Power BI Project (.pbip with TMDL semantic model and PBIR visuals) reading the local canonical dataset.Decouples Power BI from live Hub refreshes so models are diffable in Git and reproducible offline.

Supported Engagement & Platform Modes

  • Desktop Platforms: Physical Windows PCs (physical), Citrix CVAD/DaaS (citrix), Omnissa Horizon (horizon), Windows Server RDS (rds), Azure Virtual Desktop (avd), and Windows 365 (w365).
  • Engagement Types:
    • current-state: Scope, collection quality, executive summary, SpotCheck findings, hardware/OS inventory, application footprint, right-sizing tiers.
    • performance: Time-of-day hotspot heatmaps by business hour, top offending machines, CPU/disk/memory queue diagnostics, and logon breakdown.
    • sizing: Right-sizing SKU distribution based on business-hour p95 CPU and memory demand.
    • migration-baseline: Before-and-after KPI deltas comparing legacy VDI/RDS against target AVD or Windows 365 pools via baseline.json.

Public Reference Material in liquidware-ninja

While the ux-assess codebase remains private, its full methodology, SpotCheck threshold catalog, API reference, and Power BI architecture are documented openly in this repository:

SpotCheck & Canonical Dataset KB Stratusphere API KB Field Note: SpotCheck to Power BI PBIP